Hugging Face's logo

Datasets:
midah
/
hf-model-trees

Modalities:
Tabular
Text
Formats:
csv
Size:
< 1K
Libraries:
Datasets
Dask
Dataset card Data Studio Files
xet
 Community
1
midah commited on
Commit
9e81604
·
verified ·
1 Parent(s): 1a80214

Add files using upload-large-folder tool

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. 7th_Layer_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv +40 -0
  2. ACertainModel_finetunes_20250427_003734.csv_finetunes_20250427_003734.csv +91 -0
  3. Baichuan-13B-Base_finetunes_20250426_221535.csv_finetunes_20250426_221535.csv +142 -0
  4. Baichuan-13B-Chat_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv +195 -0
  5. Bio_ClinicalBERT_finetunes_20250426_121513.csv_finetunes_20250426_121513.csv +0 -0
  6. ChatTTS_finetunes_20250424_193500.csv_finetunes_20250424_193500.csv +56 -0
  7. ChilloutMix_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv +2 -0
  8. CodeLlama-70b-hf_finetunes_20250426_121513.csv_finetunes_20250426_121513.csv +145 -0
  9. DeBERTa-v3-base-mnli-fever-anli_finetunes_20250426_215237.csv_finetunes_20250426_215237.csv +0 -0
  10. DeepSeek-R1-GGUF_finetunes_20250424_223250.csv_finetunes_20250424_223250.csv +442 -0
  11. Flux-Super-Realism-LoRA_finetunes_20250426_215237.csv_finetunes_20250426_215237.csv +151 -0
  12. GR00T-N1-2B_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv +16 -0
  13. Hunyuan3D-2_finetunes_20250424_223250.csv_finetunes_20250424_223250.csv +459 -0
  14. Idefics3-8B-Llama3_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv +607 -0
  15. InstantID_finetunes_20250425_041137.csv_finetunes_20250425_041137.csv +129 -0
  16. InternVL2-Llama3-76B_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv +652 -0
  17. Janus-Pro-1B_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv +120 -0
  18. Janus-Pro-7B_finetunes_20250422_225821.csv +0 -0
  19. LLaMA-7B_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv +60 -0
  20. Llama-2-7b-hf_finetunes_20250427_003734.csv_finetunes_20250427_003734.csv +0 -0
  21. Llama-3-8B-Instruct-262k_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv +0 -0
  22. Llama-3-8B-Lexi-Uncensored_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv +273 -0
  23. MagicAnimate_finetunes_20250426_221535.csv_finetunes_20250426_221535.csv +8 -0
  24. MeloTTS-English_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv +205 -0
  25. MiniCPM-Llama3-V-2_5-gguf_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv +15 -0
  26. MiniCPM-V-2_6-gguf_finetunes_20250427_003734.csv_finetunes_20250427_003734.csv +49 -0
  27. Molmo-7B-D-0924_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv +211 -0
  28. Nemotron-4-340B-Instruct_finetunes_20250425_125929.csv_finetunes_20250425_125929.csv +0 -0
  29. Phi-3-mini-128k-instruct_finetunes_20250424_223250.csv_finetunes_20250424_223250.csv +0 -0
  30. Phi-4-mini-instruct_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv +0 -0
  31. PixArt-XL-2-1024-MS_finetunes_20250426_215237.csv_finetunes_20250426_215237.csv +329 -0
  32. SD_PixelArt_SpriteSheet_Generator_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv +60 -0
  33. SFR-Embedding-Mistral_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv +0 -0
  34. SSD-1B_finetunes_20250424_223250.csv_finetunes_20250424_223250.csv +0 -0
  35. SenseVoiceSmall_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv +235 -0
  36. SmolDocling-256M-preview_finetunes_20250424_223250.csv_finetunes_20250424_223250.csv +597 -0
  37. SmolLM-135M_finetunes_20250426_215237.csv_finetunes_20250426_215237.csv +0 -0
  38. UAE-Large-V1_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv +0 -0
  39. Yarn-Mistral-7b-128k_finetunes_20250425_143346.csv_finetunes_20250425_143346.csv +344 -0
  40. YuE-s1-7B-anneal-en-cot_finetunes_20250426_014322.csv_finetunes_20250426_014322.csv +133 -0
  41. adetailer_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv +105 -0
  42. bark-small_finetunes_20250426_215237.csv_finetunes_20250426_215237.csv +0 -0
  43. bert-base-NER_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv +0 -0
  44. chatglm-6b_finetunes_20250424_145241.csv_finetunes_20250424_145241.csv +164 -0
  45. codebert-base_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv +0 -0
  46. colpali_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv +154 -0
  47. doll-likeness-series_finetunes_20250426_014322.csv_finetunes_20250426_014322.csv +186 -0
  48. elden-ring-diffusion_finetunes_20250426_121513.csv_finetunes_20250426_121513.csv +2 -0
  49. falcon-40b-instruct_finetunes_20250424_223250.csv_finetunes_20250424_223250.csv +389 -0
  50. flan-t5-large_finetunes_20250424_223250.csv_finetunes_20250424_223250.csv +0 -0
7th_Layer_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ syaimu/7th_Layer,"---
3
+ license: other
4
+ ---
5
+
6
+ ## / 7th Layer /
7
+
8
+ <img src=""https://i.imgur.com/MjnczlB.png"" width=""1700"" height="""">
9
+
10
+ # (Important Notice:1.6)
11
+
12
+ default CFG Scale : 7 ±5
13
+
14
+ default Sampler : DPM++ 2M Karras
15
+
16
+ default Steps : 25
17
+
18
+ Negative prompt : (worst quality:1.4), (low quality:1.4) , (monochrome:1.1),
19
+ # Don't write a lot of ""Negative prompt"".
20
+ <img src=""https://i.imgur.com/tE3PUBi.png"" width=""480"" height="""">
21
+
22
+
23
+ ## Test Model https://huggingface.co/syaimu/7th_test
24
+
25
+ <img src=""https://i.imgur.com/0xKIUvL.jpg"" width=""1700"" height="""">
26
+ <img src=""https://i.imgur.com/lFZAYVv.jpg"" width=""1700"" height="""">
27
+ <img src=""https://i.imgur.com/4IYqlYq.jpg"" width=""1700"" height="""">
28
+ <img src=""https://i.imgur.com/v2pn57R.jpg"" width=""1700"" height="""">
29
+
30
+ # 7th_anime_v2.5_B → 7th_anime_v2_G
31
+ <img src=""https://i.imgur.com/K3o28Ci.jpg"" width=""1700"" height="""">
32
+ <img src=""https://i.imgur.com/Bzywbkp.jpg"" width=""1700"" height="""">
33
+
34
+ # other
35
+ <img src=""https://i.imgur.com/oCZyzdA.jpg"" width=""1700"" height="""">
36
+ <img src=""https://i.imgur.com/sAw842D.jpg"" width=""1700"" height="""">
37
+ <img src=""https://i.imgur.com/lzuYVh0.jpg"" width=""1700"" height="""">
38
+ <img src=""https://i.imgur.com/dOXsoeg.jpg"" width=""1700"" height="""">
39
+
40
+ ","{""id"": ""syaimu/7th_Layer"", ""author"": ""syaimu"", ""sha"": ""45e8fa0c30f2b3090aa4dfc6b05f7e14918cebf6"", ""last_modified"": ""2023-10-06 13:47:47+00:00"", ""created_at"": ""2022-12-27 06:10:12+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 629, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""license:other"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""license: other"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_SemiR_v3.2/7th_SemiR_v3A.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_SemiR_v3.2/7th_SemiR_v3B.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_SemiR_v3.2/7th_SemiR_v3C.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_alpha_v4/7th_anime_v4A.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_alpha_v4/7th_anime_v4B.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v1/7th_anime_v1.1.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v1/7th_anime_v1.1.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v2/7th_anime_v2_A.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v2/7th_anime_v2_A.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v2/7th_anime_v2_B-fix.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v2/7th_anime_v2_B.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v2/7th_anime_v2_B.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v2/7th_anime_v2_C.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v2/7th_anime_v2_C.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v2/7th_anime_v2_G.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v2/7th_anime_v2_G.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v3/7th_anime_v3_A.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v3/7th_anime_v3_A.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v3/7th_anime_v3_B.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v3/7th_anime_v3_B.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v3/7th_anime_v3_C.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_anime_v3/7th_anime_v3_C.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_layer/Abyss_7th_layer.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='7th_layer/abyss_7th_layerG1.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [""b3xxf21f/A3Private"", ""Phasmanta/Space2""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-10-06 13:47:47+00:00"", ""cardData"": ""license: other"", ""transformersInfo"": null, ""_id"": ""63aa8c448949ceef24a2791b"", ""modelId"": ""syaimu/7th_Layer"", ""usedStorage"": 118530128971}",0,,0,,0,,0,,0,"Phasmanta/Space2, b3xxf21f/A3Private, huggingface/InferenceSupport/discussions/new?title=syaimu/7th_Layer&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bsyaimu%2F7th_Layer%5D(%2Fsyaimu%2F7th_Layer)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A",3
ACertainModel_finetunes_20250427_003734.csv_finetunes_20250427_003734.csv ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ JosephusCheung/ACertainModel,"---
3
+ language:
4
+ - en
5
+ license: creativeml-openrail-m
6
+ tags:
7
+ - stable-diffusion
8
+ - stable-diffusion-diffusers
9
+ - text-to-image
10
+ - diffusers
11
+ inference: true
12
+ widget:
13
+ - text: ""masterpiece, best quality, 1girl, brown hair, green eyes, colorful, autumn, cumulonimbus clouds, lighting, blue sky, falling leaves, garden""
14
+ example_title: ""example 1girl""
15
+ - text: ""masterpiece, best quality, 1boy, brown hair, green eyes, colorful, autumn, cumulonimbus clouds, lighting, blue sky, falling leaves, garden""
16
+ example_title: ""example 1boy""
17
+ ---
18
+
19
+ # ACertainModel
20
+
21
+ **Try full functions with Google Colab free T4** [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1ldhBc70wvuvkp4Af_vNTzTfBXwpf_cH5?usp=sharing)
22
+
23
+ Check Twitter [#ACertainModel](https://twitter.com/hashtag/ACertainModel) for community artworks
24
+
25
+ Welcome to ACertainModel - a latent diffusion model for weebs. This model is intended to produce high-quality, highly detailed anime style pictures with just a few prompts. Like other anime-style Stable Diffusion models, it also supports danbooru tags, including artists, to generate images.
26
+
27
+ Since I noticed that the laion-aesthetics introduced in the Stable-Diffusion-v-1-4 checkpoint hindered finetuning anime style illustration generation model, Dreambooth was used to finetune some tags separately to make it closer to what it was in SD1.2. To avoid overfitting and possible language drift, I added a huge amount of auto-generated pictures from a single word prompt to the training set, using models that are popular in the community such as Anything-3.0, together with partially manual selected full-danbooru images within a year, for further native training. I am also aware of a method of [LoRA](https://arxiv.org/abs/2106.09685), with a similar idea, finetuning attention layer solely, to have better performance on eyes, hands, and other details.
28
+
29
+ For copyright compliance and technical experiment, it was trained from few artist images directly. It was trained on Dreambooth with pictures generated from several popular diffusion models in the community. The checkpoint was initialized with the weights of a Stable Diffusion Model and subsequently fine-tuned for 2K GPU hours on V100 32GB and 600 GPU hours on A100 40GB at 512P dynamic aspect ratio resolution with a certain ratio of unsupervised auto-generated images from several popular diffusion models in the community with some Textual Inversions and Hypernetworks. We do know some tricks on xformers and 8-bit optimization, but we didn't use any of them for better quality and stability. Up to 15 branches are trained simultaneously, cherry-picking about every 20,000 steps.
30
+
31
+ e.g. **_masterpiece, best quality, 1girl, brown hair, green eyes, colorful, autumn, cumulonimbus clouds, lighting, blue sky, falling leaves, garden_**
32
+
33
+ ## About online preview with Hosted inference API, also generation with this model
34
+
35
+ Parameters are not allowed to be modified, as it seems that it is generated with *Clip skip: 1*, for better performance, it is strongly recommended to use *Clip skip: 2* instead.
36
+
37
+ Here is an example of inference settings, if it is applicable with you on your own server: *Steps: 28, Sampler: Euler a, CFG scale: 11, Clip skip: 2*.
38
+
39
+ ## 🧨 Diffusers
40
+
41
+ This model can be used just like any other Stable Diffusion model. For more information,
42
+ please have a look at the [Stable Diffusion](https://huggingface.co/docs/diffusers/api/pipelines/stable_diffusion).
43
+
44
+ You can also export the model to [ONNX](https://huggingface.co/docs/diffusers/optimization/onnx), [MPS](https://huggingface.co/docs/diffusers/optimization/mps) and/or FLAX/JAX.
45
+
46
+ ```python
47
+ from diffusers import StableDiffusionPipeline
48
+ import torch
49
+
50
+ model_id = ""JosephusCheung/ACertainModel""
51
+ branch_name= ""main""
52
+
53
+ pipe = StableDiffusionPipeline.from_pretrained(model_id, revision=branch_name, torch_dtype=torch.float16)
54
+ pipe = pipe.to(""cuda"")
55
+
56
+ prompt = ""pikachu""
57
+ image = pipe(prompt).images[0]
58
+
59
+ image.save(""./pikachu.png"")
60
+ ```
61
+
62
+ ## Examples
63
+
64
+ Below are some examples of images generated using this model, with better performance on framing and hand gestures, as well as moving objects, comparing to other analogues:
65
+
66
+ **Anime Girl:**
67
+ ![Anime Girl](https://huggingface.co/JosephusCheung/ACertainModel/resolve/main/samples/sample-1girl.png)
68
+ ```
69
+ 1girl, brown hair, green eyes, colorful, autumn, cumulonimbus clouds, lighting, blue sky, falling leaves, garden
70
+ Steps: 28, Sampler: Euler a, CFG scale: 11, Seed: 114514, Clip skip: 2
71
+ ```
72
+ **Anime Boy:**
73
+ ![Anime Boy](https://huggingface.co/JosephusCheung/ACertainModel/resolve/main/samples/sample-1boy.png)
74
+ ```
75
+ 1boy, brown hair, green eyes, colorful, autumn, cumulonimbus clouds, lighting, blue sky, falling leaves, garden
76
+ Steps: 28, Sampler: Euler a, CFG scale: 11, Seed: 114514, Clip skip: 2
77
+ ```
78
+
79
+ ## License
80
+
81
+ This model is open access and available to all, with a CreativeML OpenRAIL-M license further specifying rights and usage.
82
+ The CreativeML OpenRAIL License specifies:
83
+
84
+ 1. You can't use the model to deliberately produce nor share illegal or harmful outputs or content
85
+ 2. The authors claims no rights on the outputs you generate, you are free to use them and are accountable for their use which must not go against the provisions set in the license
86
+ 3. You may re-distribute the weights and use the model commercially and/or as a service. If you do, please be aware you have to include the same use restrictions as the ones in the license and share a copy of the CreativeML OpenRAIL-M to all your users (please read the license entirely and carefully)
87
+ [Please read the full license here](https://huggingface.co/spaces/CompVis/stable-diffusion-license)
88
+
89
+ ## Is it a NovelAI based model? What is the relationship with SD1.2 and SD1.4?
90
+
91
+ See [ASimilarityCalculatior](https://huggingface.co/JosephusCheung/ASimilarityCalculatior)","{""id"": ""JosephusCheung/ACertainModel"", ""author"": ""JosephusCheung"", ""sha"": ""02fc9a2dccf7ebce834fb17f53a304dc77d679ba"", ""last_modified"": ""2022-12-20 03:16:49+00:00"", ""created_at"": ""2022-12-12 17:40:00+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 426, ""downloads_all_time"": null, ""likes"": 159, ""library_name"": ""diffusers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""diffusers"", ""stable-diffusion"", ""stable-diffusion-diffusers"", ""text-to-image"", ""en"", ""arxiv:2106.09685"", ""doi:10.57967/hf/0196"", ""license:creativeml-openrail-m"", ""autotrain_compatible"", ""endpoints_compatible"", ""diffusers:StableDiffusionPipeline"", ""region:us""], ""pipeline_tag"": ""text-to-image"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""language:\n- en\nlicense: creativeml-openrail-m\ntags:\n- stable-diffusion\n- stable-diffusion-diffusers\n- text-to-image\n- diffusers\ninference: true\nwidget:\n- text: masterpiece, best quality, 1girl, brown hair, green eyes, colorful, autumn,\n cumulonimbus clouds, lighting, blue sky, falling leaves, garden\n example_title: example 1girl\n- text: masterpiece, best quality, 1boy, brown hair, green eyes, colorful, autumn,\n cumulonimbus clouds, lighting, blue sky, falling leaves, garden\n example_title: example 1boy"", ""widget_data"": [{""text"": ""masterpiece, best quality, 1girl, brown hair, green eyes, colorful, autumn, cumulonimbus clouds, lighting, blue sky, falling leaves, garden"", ""example_title"": ""example 1girl""}, {""text"": ""masterpiece, best quality, 1boy, brown hair, green eyes, colorful, autumn, cumulonimbus clouds, lighting, blue sky, falling leaves, garden"", ""example_title"": ""example 1boy""}], ""model_index"": null, ""config"": {""diffusers"": {""_class_name"": ""StableDiffusionPipeline""}}, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ACertainModel-half.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ACertainModel.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ACertainModel_on_SD_WEBUI.ipynb', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='feature_extractor/preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model_index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='safety_checker/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='safety_checker/pytorch_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='samples/anything3-sample-1boy.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='samples/anything3-sample-1girl.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='samples/sample-1boy.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='samples/sample-1girl.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='scheduler/scheduler_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='text_encoder/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='text_encoder/pytorch_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer/merges.txt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer/special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer/tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer/vocab.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='unet/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='unet/diffusion_pytorch_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='vae/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='vae/diffusion_pytorch_model.bin', size=None, blob_id=None, lfs=None)""], ""spaces"": [""Yntec/ToyWorld"", ""Yntec/PrintingPress"", ""Nymbo/image_gen_supaqueue"", ""ennov8ion/3dart-Models"", ""phenixrhyder/NSFW-ToyWorld"", ""Yntec/blitz_diffusion"", ""sanaweb/text-to-image"", ""Vedits/6x_Image_diffusion"", ""John6666/Diffusion80XX4sg"", ""ennov8ion/comicbook-models"", ""John6666/PrintingPress4"", ""PeepDaSlan9/B2BMGMT_Diffusion60XX"", ""Daniela-C/6x_Image_diffusion"", ""phenixrhyder/PrintingPress"", ""John6666/hfd_test_nostopbutton"", ""mindtube/Diffusion50XX"", ""TheKitten/Fast-Images-Creature"", ""Nymbo/Diffusion80XX4sg"", ""kaleidoskop-hug/PrintingPress"", ""ennov8ion/stablediffusion-models"", ""John6666/ToyWorld4"", ""grzegorz2047/fast_diffusion"", ""Alfasign/dIFFU"", ""Nymbo/PrintingPress"", ""Rifd/Sdallmodels"", ""John6666/Diffusion80XX4g"", ""NativeAngels/HuggingfaceDiffusion"", ""ennov8ion/Scifi-Models"", ""ennov8ion/semirealistic-models"", ""ennov8ion/FantasyArt-Models"", ""ennov8ion/dreamlike-models"", ""noes14155/img_All_models"", ""ennov8ion/500models"", ""AnimeStudio/anime-models"", ""John6666/Diffusion80XX4"", ""K00B404/HuggingfaceDiffusion_custom"", ""John6666/blitz_diffusion4"", ""John6666/blitz_diffusion_builtin"", ""RhythmRemix14/PrintingPressDx"", ""sohoso/PrintingPress"", ""NativeAngels/ToyWorld"", ""mindtube/maximum_multiplier_places"", ""animeartstudio/AnimeArtmodels2"", ""animeartstudio/AnimeModels"", ""Binettebob22/fast_diffusion2"", ""pikto/Elite-Scifi-Models"", ""PixelistStudio/3dart-Models"", ""devmiles/zexxiai"", ""Nymbo/Diffusion60XX"", ""TheKitten/Images"", ""ennov8ion/anime-models"", ""jordonpeter01/Diffusion70"", ""mimimibimimimi/ACertainModel"", ""Phasmanta/Space2"", ""ennov8ion/Landscapes-models"", ""sohoso/anime348756"", ""ucmisanddisinfo/thisApp"", ""johann22/chat-diffusion"", ""K00B404/generate_many_models"", ""manivannan7gp/Words2Image"", ""ennov8ion/art-models"", ""ennov8ion/photo-models"", ""ennov8ion/art-multi"", ""vih-v/x_mod"", ""NativeAngels/blitz_diffusion"", ""NativeAngels/PrintingPress4"", ""NativeAngels/PrintingPress"", ""dehua68/ToyWorld"", ""burman-ai/Printing-Press"", ""sk16er/ghibli_creator"", ""fo-atccb/ACertainModel"", ""ecody726/JosephusCheung-ACertainModel"", ""cap1145/JosephusCheung-ACertainModel"", ""ennov8ion/abstractart-models"", ""ennov8ion/Scifiart-Models"", ""ennov8ion/interior-models"", ""ennov8ion/room-interior-models"", ""animeartstudio/AnimeArtModels1"", ""Yntec/top_100_diffusion"", ""AIlexDev/Diffusion60XX"", ""flatindo/img_All_models"", ""flatindo/all-models"", ""flatindo/all-models-v1"", ""johann22/chat-diffusion-describe"", ""wideprism/Ultimate-Model-Collection"", ""GAIneZis/FantasyArt-Models"", ""TheMaisk/Einfach.ImageAI"", ""ennov8ion/picasso-diffusion"", ""K00B404/stablediffusion-portal"", ""ennov8ion/anime-new-models"", ""ennov8ion/anime-multi-new-models"", ""ennov8ion/photo-multi"", ""ennov8ion/anime-multi"", ""Ashrafb/comicbook-models"", ""sohoso/architecture"", ""K00B404/image_gen_supaqueue_game_assets"", ""GhadaSaylami/text-to-image"", ""Geek7/mdztxi"", ""Geek7/mdztxi2"", ""NativeAngels/Diffusion80XX4sg""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2022-12-20 03:16:49+00:00"", ""cardData"": ""language:\n- en\nlicense: creativeml-openrail-m\ntags:\n- stable-diffusion\n- stable-diffusion-diffusers\n- text-to-image\n- diffusers\ninference: true\nwidget:\n- text: masterpiece, best quality, 1girl, brown hair, green eyes, colorful, autumn,\n cumulonimbus clouds, lighting, blue sky, falling leaves, garden\n example_title: example 1girl\n- text: masterpiece, best quality, 1boy, brown hair, green eyes, colorful, autumn,\n cumulonimbus clouds, lighting, blue sky, falling leaves, garden\n example_title: example 1boy"", ""transformersInfo"": null, ""_id"": ""6397677008a51789e4b50b25"", ""modelId"": ""JosephusCheung/ACertainModel"", ""usedStorage"": 12711403487}",0,,0,,0,,0,,0,"CompVis/stable-diffusion-license, Daniela-C/6x_Image_diffusion, John6666/Diffusion80XX4sg, John6666/PrintingPress4, John6666/ToyWorld4, John6666/hfd_test_nostopbutton, Nymbo/image_gen_supaqueue, PeepDaSlan9/B2BMGMT_Diffusion60XX, Yntec/PrintingPress, Yntec/ToyWorld, Yntec/blitz_diffusion, huggingface/InferenceSupport/discussions/new?title=JosephusCheung/ACertainModel&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BJosephusCheung%2FACertainModel%5D(%2FJosephusCheung%2FACertainModel)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, kaleidoskop-hug/PrintingPress, phenixrhyder/NSFW-ToyWorld",14
Baichuan-13B-Base_finetunes_20250426_221535.csv_finetunes_20250426_221535.csv ADDED
@@ -0,0 +1,142 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ baichuan-inc/Baichuan-13B-Base,"---
3
+ language:
4
+ - zh
5
+ - en
6
+ pipeline_tag: text-generation
7
+ inference: false
8
+ ---
9
+ # Baichuan-13B-Base
10
+
11
+ <!-- Provide a quick summary of what the model is/does. -->
12
+
13
+ ## 介绍
14
+ Baichuan-13B-Base为Baichuan-13B系列模型中的预训练版本，经过对齐后的模型可见[Baichuan-13B-Chat](https://huggingface.co/baichuan-inc/Baichuan-13B-Chat)。
15
+
16
+ [Baichuan-13B](https://github.com/baichuan-inc/Baichuan-13B) 是由百川智能继 [Baichuan-7B](https://github.com/baichuan-inc/baichuan-7B) 之后开发的包含 130 亿参数的开源可商用的大规模语言模型，在权威的中文和英文 benchmark 上均取得同尺寸最好的效果。本次发布包含有预训练 ([Baichuan-13B-Base](https://huggingface.co/baichuan-inc/Baichuan-13B-Base)) 和对齐 ([Baichuan-13B-Chat](https://huggingface.co/baichuan-inc/Baichuan-13B-Chat)) 两个版本。Baichuan-13B 有如下几个特点：
17
+
18
+ 1. **更大尺寸、更多数据**：Baichuan-13B 在 [Baichuan-7B](https://github.com/baichuan-inc/baichuan-7B) 的基础上进一步扩大参数量到 130 亿，并且在高质量的语料上训练了 1.4 万亿 tokens，超过 LLaMA-13B 40%，是当前开源 13B 尺寸下训练数据量最多的模型。支持中英双语，使用 ALiBi 位置编码，上下文窗口长度为 4096。
19
+ 2. **同时开源预训练和对齐模型**：预训练模型是适用开发者的“基座”，而广大普通用户对有对话功能的对齐模型具有更强的需求。因此本次开源我们同时发布了对齐模型（Baichuan-13B-Chat），具有很强的对话能力，开箱即用，几行代码即可简单的部署。
20
+ 3. **更高效的推理**：为了支持更广大用户的使用，我们本次同时开源了 int8 和 int4 的量化版本，相对非量化版本在几乎没有效果损失的情况下大大降低了部署的机器资源门槛，可以部署在如 Nvidia 3090 这样的消费级显卡上。
21
+ 4. **开源免费可商用**：Baichuan-13B 不仅对学术研究完全开放，开发者也仅需邮件申请并获得官方商用许可后，即可以免费商用。
22
+ 5.
23
+ Baichuan-13B-Base is the pre-training version in the Baichuan-13B series of models, and the aligned model can be found at [Baichuan-13B-Chat](https://huggingface.co/baichuan-inc/Baichuan-13B-Chat).
24
+
25
+ [Baichuan-13B](https://github.com/baichuan-inc/Baichuan-13B) is an open-source, commercially usable large-scale language model developed by Baichuan Intelligence, following [Baichuan-7B](https://github.com/baichuan-inc/baichuan-7B). With 13 billion parameters, it achieves the best performance in standard Chinese and English benchmarks among models of its size. This release includes two versions: pre-training (Baichuan-13B-Base) and alignment (Baichuan-13B-Chat). Baichuan-13B has the following features:
26
+
27
+ 1. **Larger size, more data**: Baichuan-13B further expands the parameter volume to 13 billion based on [Baichuan-7B](https://github.com/baichuan-inc/baichuan-7B), and has trained 1.4 trillion tokens on high-quality corpora, exceeding LLaMA-13B by 40%. It is currently the model with the most training data in the open-source 13B size. It supports both Chinese and English, uses ALiBi position encoding, and has a context window length of 4096.
28
+ 2. **Open-source pre-training and alignment models simultaneously**: The pre-training model is a ""base"" suitable for developers, while the general public has a stronger demand for alignment models with dialogue capabilities. Therefore, in this open-source release, we also released the alignment model (Baichuan-13B-Chat), which has strong dialogue capabilities and is ready to use. It can be easily deployed with just a few lines of code.
29
+ 3. **More efficient inference**: To support a wider range of users, we have open-sourced the INT8 and INT4 quantized versions. The model can be conveniently deployed on consumer GPUs like the Nvidia 3090 with almost no performance loss.
30
+ 4. **Open-source, free, and commercially usable**: Baichuan-13B is not only fully open to academic research, but developers can also use it for free commercially after applying for and receiving official commercial permission via email.
31
+
32
+
33
+ ## 模型详情
34
+
35
+ ### 模型描述
36
+
37
+ <!-- Provide a longer summary of what this model is. -->
38
+
39
+ - **Developed by:** 百川智能(Baichuan Intelligent Technology)
40
+ - **Email**: opensource@baichuan-inc.com
41
+ - **Language(s) (NLP):** Chinese/English
42
+ - **License:** 【Community License for Baichuan-13B Model】([ZH](Baichuan-13B%20模型社区许可协议.pdf)|
43
+ [EN](Community%20License%20for%20Baichuan-13B%20Model.pdf))
44
+
45
+ **商业用途(For commercial use):** 请通过 [Email](mailto:opensource@baichuan-inc.com) 联系申请书面授权。(Contact us via [Email](mailto:opensource@baichuan-inc.com) above to apply for written authorization.)
46
+
47
+ ### 模型结构
48
+
49
+ <!-- Provide the basic links for the model. -->
50
+
51
+ 整体模型基于Baichuan-7B，为了获得更好的推理性能，Baichuan-13B 使用了 ALiBi 线性偏置技术，相对于 Rotary Embedding 计算量更小，对推理性能有显著提升；与标准的 LLaMA-13B 相比，生成 2000 个 tokens 的平均推理速度 (tokens/s)，实测提升 31.6%：
52
+
53
+ | Model | tokens/s |
54
+ |-------------|----------|
55
+ | LLaMA-13B | 19.4 |
56
+ | Baichuan-13B| 25.4 |
57
+
58
+ 具体参数和见下表
59
+ | 模型名称 | 隐含层维度 | 层数 | 头数 |词表大小 | 总参数量 | 训练数据（tokens） | 位置编码 | 最大长度 |
60
+ |-------------------------|-------|------------|------------|-----------------|--------|--------|----------------|---------|
61
+ | Baichuan-7B | 4,096 | 32 | 32 | 64,000 | 7,000,559,616 | 1.2万亿 | [RoPE](https://arxiv.org/abs/2104.09864) | 4,096 |
62
+ | Baichuan-13B | 5,120 | 40 | 40 | 64,000 | 13,264,901,120 | 1.4万亿 | [ALiBi](https://arxiv.org/abs/2108.12409) | 4,096
63
+
64
+ The overall model is based on Baichuan-7B. In order to achieve better inference performance, Baichuan-13B uses ALiBi linear bias technology, which has a smaller computational load compared to Rotary Embedding, and significantly improves inference performance. Compared with the standard LLaMA-13B, the average inference speed (tokens/s) for generating 2000 tokens has been tested to increase by 31.6%:
65
+
66
+ | Model | tokens/s |
67
+ |-------------|----------|
68
+ | LLaMA-13B | 19.4 |
69
+ | Baichuan-13B| 25.4 |
70
+
71
+ The specific parameters are as follows:
72
+ | Model Name | Hidden Size | Num Layers | Num Attention Heads |Vocab Size | Total Params | Training Dats（tokens） | Position Embedding | Max Length |
73
+ |-------------------------|-------|------------|------------|-----------------|--------|--------|----------------|---------|
74
+ | Baichuan-7B | 4,096 | 32 | 32 | 64,000 | 7,000,559,616 | 1.2万亿 | [RoPE](https://arxiv.org/abs/2104.09864) | 4,096 |
75
+ | Baichuan-13B | 5,120 | 40 | 40 | 64,000 | 13,264,901,120 | 1.4万亿 | [ALiBi](https://arxiv.org/abs/2108.12409) | 4,096
76
+
77
+ ### 免责声明
78
+
79
+ 我们在此声明，我们的开发团队并未基于 Baichuan-13B 模型开发任何应用，无论是在 iOS、Android、网页或任何其他平台。我们强烈呼吁所有使用者，不要利用 Baichuan-13B 模型进行任何危害国家社会安全或违法的活动。另外，我们也要求使用者不要将 Baichuan-13B 模型用于未经适当安全审查和备案的互联网服务。我们希望所有的使用者都能遵守这个原则，确保科技的发展能在规范和合法的环境下进行。
80
+
81
+ 我们已经尽我们所能，来确保模型训练过程中使用的数据的合规性。然而，尽管我们已经做出了巨大的努力，但由于模型和数据的复杂性，仍有可能存在一些无法预见的问题。因此，如果由于使用 Baichuan-13B 开源模型而导致的任何问题，包括但不限于数据安全问题、公共舆论风险，或模型被误导、滥用、传播或不当利用所带来的任何风险和问题，我们将不承担任何责任。
82
+
83
+ We hereby declare that our development team has not developed any applications based on the Baichuan-13B model, whether on iOS, Android, the web, or any other platform. We strongly urge all users not to use the Baichuan-13B model for any activities that harm national social security or are illegal. In addition, we also ask users not to use the Baichuan-13B model for internet services that have not undergone appropriate security review and filing. We hope that all users will adhere to this principle to ensure that technological development takes place in a regulated and legal environment.
84
+
85
+ We have done our utmost to ensure the compliance of the data used in the model training process. However, despite our great efforts, due to the complexity of the model and data, there may still be some unforeseen issues. Therefore, we will not take any responsibility for any issues arising from the use of the Baichuan-13B open-source model, including but not limited to data security issues, public opinion risks, or any risks and problems arising from the model being misled, misused, disseminated, or improperly exploited.
86
+
87
+ ## 训练详情
88
+
89
+ 训练具体设置参见[Baichuan-13B](https://github.com/baichuan-inc/Baichuan-13B)。
90
+
91
+ For specific training settings, please refer to [Baichuan-13B](https://github.com/baichuan-inc/Baichuan-13B).
92
+
93
+ ## 测评结果
94
+
95
+ ### [C-Eval](https://cevalbenchmark.com/index.html#home)
96
+
97
+ | Model 5-shot | STEM | Social Sciences | Humanities | Others | Average |
98
+ |-------------------------|:-----:|:---------------:|:----------:|:------:|:-------:|
99
+ | Baichuan-7B | 38.2 | 52.0 | 46.2 | 39.3 | 42.8 |
100
+ | Chinese-Alpaca-Plus-13B | 35.2 | 45.6 | 40.0 | 38.2 | 38.8 |
101
+ | Vicuna-13B | 30.5 | 38.2 | 32.5 | 32.5 | 32.8 |
102
+ | Chinese-LLaMA-Plus-13B | 30.3 | 38.0 | 32.9 | 29.1 | 32.1 |
103
+ | Ziya-LLaMA-13B-Pretrain | 27.6 | 34.4 | 32.0 | 28.6 | 30.0 |
104
+ | LLaMA-13B | 27.0 | 33.6 | 27.7 | 27.6 | 28.5 |
105
+ | moss-moon-003-base (16B)| 27.0 | 29.1 | 27.2 | 26.9 | 27.4 |
106
+ | **Baichuan-13B-Base** | **45.9** | **63.5** | **57.2** | **49.3** | **52.4** |
107
+ | **Baichuan-13B-Chat** | **43.7** | **64.6** | **56.2** | **49.2** | **51.5** |
108
+
109
+
110
+ ### [MMLU](https://arxiv.org/abs/2009.03300)
111
+
112
+ | Model 5-shot | STEM | Social Sciences | Humanities | Others | Average |
113
+ |-------------------------|:-----:|:---------------:|:----------:|:------:|:-------:|
114
+ | Vicuna-13B | 40.4 | 60.5 | 49.5 | 58.4 | 52.0 |
115
+ | LLaMA-13B | 36.1 | 53.0 | 44.0 | 52.8 | 46.3 |
116
+ | Chinese-Alpaca-Plus-13B | 36.9 | 48.9 | 40.5 | 50.5 | 43.9 |
117
+ | Ziya-LLaMA-13B-Pretrain | 35.6 | 47.6 | 40.1 | 49.4 | 42.9 |
118
+ | Baichuan-7B | 35.6 | 48.9 | 38.4 | 48.1 | 42.3 |
119
+ | Chinese-LLaMA-Plus-13B | 33.1 | 42.8 | 37.0 | 44.6 | 39.2 |
120
+ | moss-moon-003-base (16B)| 22.4 | 22.8 | 24.2 | 24.4 | 23.6 |
121
+ | **Baichuan-13B-Base** | **41.6** | **60.9** | **47.4** | **58.5** | **51.6** |
122
+ | **Baichuan-13B-Chat** | **40.9** | **60.9** | **48.8** | **59.0** | **52.1** |
123
+ > 说明：我们采用了 MMLU 官方的[评测方案](https://github.com/hendrycks/test)。
124
+
125
+ ### [CMMLU](https://github.com/haonan-li/CMMLU)
126
+
127
+ | Model 5-shot | STEM | Humanities | Social Sciences | Others | China Specific | Average |
128
+ |-------------------------|:-----:|:----------:|:---------------:|:------:|:--------------:|:-------:|
129
+ | Baichuan-7B | 34.4 | 47.5 | 47.6 | 46.6 | 44.3 | 44.0 |
130
+ | Vicuna-13B | 31.8 | 36.2 | 37.6 | 39.5 | 34.3 | 36.3 |
131
+ | Chinese-Alpaca-Plus-13B | 29.8 | 33.4 | 33.2 | 37.9 | 32.1 | 33.4 |
132
+ | Chinese-LLaMA-Plus-13B | 28.1 | 33.1 | 35.4 | 35.1 | 33.5 | 33.0 |
133
+ | Ziya-LLaMA-13B-Pretrain | 29.0 | 30.7 | 33.8 | 34.4 | 31.9 | 32.1 |
134
+ | LLaMA-13B | 29.2 | 30.8 | 31.6 | 33.0 | 30.5 | 31.2 |
135
+ | moss-moon-003-base (16B)| 27.2 | 30.4 | 28.8 | 32.6 | 28.7 | 29.6 |
136
+ | **Baichuan-13B-Base** | **41.7** | **61.1** | **59.8** | **59.0** | **56.4** | **55.3** |
137
+ | **Baichuan-13B-Chat** | **42.8** | **62.6** | **59.7** | **59.0** | **56.1** | **55.8** |
138
+ > 说明：CMMLU 是一个综合性的中文评估基准，专门用于评估语言模型在中文语境下的知识和推理能力。我们采用了其官方的[评测方案](https://github.com/haonan-li/CMMLU)。
139
+
140
+ ## 微信群组
141
+ ![WeChat](https://github.com/baichuan-inc/Baichuan-13B/blob/main/media/wechat.jpeg?raw=true)
142
+ ","{""id"": ""baichuan-inc/Baichuan-13B-Base"", ""author"": ""baichuan-inc"", ""sha"": ""0ef0739c7bdd34df954003ef76d80f3dabca2ff9"", ""last_modified"": ""2023-07-19 03:37:12+00:00"", ""created_at"": ""2023-07-08 16:55:46+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 418, ""downloads_all_time"": null, ""likes"": 185, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""pytorch"", ""baichuan"", ""text-generation"", ""custom_code"", ""zh"", ""en"", ""arxiv:2104.09864"", ""arxiv:2108.12409"", ""arxiv:2009.03300"", ""autotrain_compatible"", ""text-generation-inference"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""language:\n- zh\n- en\npipeline_tag: text-generation\ninference: false"", ""widget_data"": [{""text"": ""\u6211\u53eb\u6731\u5229\u5b89\uff0c\u6211\u559c\u6b22""}, {""text"": ""\u6211\u53eb\u6258\u9a6c\u65af\uff0c\u6211\u7684\u4e3b\u8981""}, {""text"": ""\u6211\u53eb\u739b\u4e3d\u4e9a\uff0c\u6211\u6700\u559c\u6b22\u7684""}, {""text"": ""\u6211\u53eb\u514b\u62c9\u62c9\uff0c\u6211\u662f""}, {""text"": ""\u4ece\u524d\uff0c""}], ""model_index"": null, ""config"": {""architectures"": [""BaichuanForCausalLM""], ""auto_map"": {""AutoConfig"": ""configuration_baichuan.BaichuanConfig"", ""AutoModelForCausalLM"": ""modeling_baichuan.BaichuanForCausalLM""}, ""model_type"": ""baichuan"", ""tokenizer_config"": {""bos_token"": {""__type"": ""AddedToken"", ""content"": ""<s>"", ""lstrip"": false, ""normalized"": true, ""rstrip"": false, ""single_word"": true}, ""eos_token"": {""__type"": ""AddedToken"", ""content"": ""</s>"", ""lstrip"": false, ""normalized"": true, ""rstrip"": false, ""single_word"": true}, ""pad_token"": {""__type"": ""AddedToken"", ""content"": ""<unk>"", ""lstrip"": false, ""normalized"": true, ""rstrip"": false, ""single_word"": true}, ""unk_token"": {""__type"": ""AddedToken"", ""content"": ""<unk>"", ""lstrip"": false, ""normalized"": true, ""rstrip"": false, ""single_word"": true}}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": ""modeling_baichuan.BaichuanForCausalLM"", ""pipeline_tag"": ""text-generation"", ""processor"": null}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Baichuan-13B \u6a21\u578b\u793e\u533a\u8bb8\u53ef\u534f\u8bae.pdf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Community License for Baichuan-13B Model.pdf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='configuration_baichuan.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='modeling_baichuan.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00001-of-00003.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00002-of-00003.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00003-of-00003.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model.bin.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='quantizer.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='requirements.txt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenization_baichuan.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.model', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [""Justinrune/LLaMA-Factory"", ""Junity/Genshin-World-Model"", ""kenken999/fastapi_django_main_live"", ""officialhimanshu595/llama-factory"", ""li-qing/FIRE"", ""tianleliphoebe/visual-arena"", ""Ashmal/MobiLlama"", ""PegaMichael/Taiwan-LLaMa2-Copy"", ""tjtanaa/chat-template-generation"", ""CaiRou-Huang/TwLLM7B-v2.0-base"", ""pallavijaini/NeuralChat-LLAMA-POC"", ""blackwingedkite/gutalk"", ""cllatMTK/Breeze"", ""WinterGYC/Genshin-World-Model-Junity"", ""gordonchan/embedding-m3e-large"", ""blackwingedkite/alpaca2_clas"", ""jaekwon/intel_cpu_chat"", ""lianglv/NeuralChat-ICX-INT4"", ""Bofeee5675/FIRE"", ""evelyn-lo/evelyn"", ""yuantao-infini-ai/demo_test"", ""zjasper666/bf16_vs_fp8"", ""martinakaduc/melt"", ""cloneQ/internLMRAG"", ""hujin0929/LlamaIndex_RAG"", ""flyfive0315/internLlamaIndex"", ""sunxiaokang/llamaindex_RAG_web"", ""kai119/llama"", ""qxy826982153/LlamaIndexRAG"", ""ilemon/Internlm2.5LLaMAindexRAG"", ""msun415/Llamole""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-07-19 03:37:12+00:00"", ""cardData"": ""language:\n- zh\n- en\npipeline_tag: text-generation\ninference: false"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": ""modeling_baichuan.BaichuanForCausalLM"", ""pipeline_tag"": ""text-generation"", ""processor"": null}, ""_id"": ""64a9951299639cc3126324e2"", ""modelId"": ""baichuan-inc/Baichuan-13B-Base"", ""usedStorage"": 53062970073}",0,,0,,0,,0,,0,"Ashmal/MobiLlama, Bofeee5675/FIRE, Junity/Genshin-World-Model, Justinrune/LLaMA-Factory, blackwingedkite/gutalk, evelyn-lo/evelyn, gordonchan/embedding-m3e-large, huggingface/InferenceSupport/discussions/new?title=baichuan-inc/Baichuan-13B-Base&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bbaichuan-inc%2FBaichuan-13B-Base%5D(%2Fbaichuan-inc%2FBaichuan-13B-Base)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, kenken999/fastapi_django_main_live, li-qing/FIRE, martinakaduc/melt, officialhimanshu595/llama-factory, tianleliphoebe/visual-arena",13
Baichuan-13B-Chat_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv ADDED
@@ -0,0 +1,195 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ baichuan-inc/Baichuan-13B-Chat,"---
3
+ language:
4
+ - zh
5
+ - en
6
+ pipeline_tag: text-generation
7
+ inference: false
8
+ ---
9
+ # Baichuan-13B-Chat
10
+
11
+ <!-- Provide a quick summary of what the model is/does. -->
12
+
13
+ ## 介绍
14
+ Baichuan-13B-Chat为Baichuan-13B系列模型中对齐后的版本，预训练模型可见[Baichuan-13B-Base](https://huggingface.co/baichuan-inc/Baichuan-13B-Base)。
15
+
16
+ [Baichuan-13B](https://github.com/baichuan-inc/Baichuan-13B) 是由百川智能继 [Baichuan-7B](https://github.com/baichuan-inc/baichuan-7B) 之后开发的包含 130 亿参数的开源可商用的大规模语言模型，在权威的中文和英文 benchmark 上均取得同尺寸最好的效果。本次发布包含有预训练 ([Baichuan-13B-Base](https://huggingface.co/baichuan-inc/Baichuan-13B-Base)) 和对齐 ([Baichuan-13B-Chat](https://huggingface.co/baichuan-inc/Baichuan-13B-Chat)) 两个版本。Baichuan-13B 有如下几个特点：
17
+
18
+ 1. **更大尺寸、更多数据**：Baichuan-13B 在 [Baichuan-7B](https://github.com/baichuan-inc/baichuan-7B) 的基础上进一步扩大参数量到 130 亿，并且在高质量的语料上训练了 1.4 万亿 tokens，超过 LLaMA-13B 40%，是当前开源 13B 尺寸下训练数据量最多的模型。支持中英双语，使用 ALiBi 位置编码，上下文窗口长度为 4096。
19
+ 2. **同时开源预训练和对齐模型**：预训练模型是适用开发者的“基座”，而广大普通用户对有对话功能的对齐模型具有更强的需求。因此本次开源我们同时发布了对齐模型（Baichuan-13B-Chat），具有很强的对话能力，开箱即用，几行代码即可简单的部署。
20
+ 3. **更高效的推理**：为了支持更广大用户的使用，我们本次同时开源了 int8 和 int4 的量化版本，相对非量化版本在几乎没有效果损失的情况下大大降低了部署的机器资源门槛，可以部署在如 Nvidia 3090 这样的消费级显卡上。
21
+ 4. **开源免费可商用**：Baichuan-13B 不仅对学术研究完全开放，开发者也仅需邮件申请并获得官方商用许可后，即可以免费商用。
22
+
23
+ Baichuan-13B-Chat is the aligned version in the Baichuan-13B series of models, and the pre-trained model can be found at [Baichuan-13B-Base](https://huggingface.co/baichuan-inc/Baichuan-13B-Base).
24
+
25
+ [Baichuan-13B](https://github.com/baichuan-inc/Baichuan-13B) is an open-source, commercially usable large-scale language model developed by Baichuan Intelligence, following [Baichuan-7B](https://github.com/baichuan-inc/baichuan-7B). With 13 billion parameters, it achieves the best performance in standard Chinese and English benchmarks among models of its size. This release includes two versions: pre-training (Baichuan-13B-Base) and alignment (Baichuan-13B-Chat). Baichuan-13B has the following features:
26
+
27
+ 1. **Larger size, more data**: Baichuan-13B further expands the parameter volume to 13 billion based on [Baichuan-7B](https://github.com/baichuan-inc/baichuan-7B), and has trained 1.4 trillion tokens on high-quality corpora, exceeding LLaMA-13B by 40%. It is currently the model with the most training data in the open-source 13B size. It supports both Chinese and English, uses ALiBi position encoding, and has a context window length of 4096.
28
+ 2. **Open-source pre-training and alignment models simultaneously**: The pre-training model is a ""base"" suitable for developers, while the general public has a stronger demand for alignment models with dialogue capabilities. Therefore, in this open-source release, we also released the alignment model (Baichuan-13B-Chat), which has strong dialogue capabilities and is ready to use. It can be easily deployed with just a few lines of code.
29
+ 3. **More efficient inference**: To support a wider range of users, we have open-sourced the INT8 and INT4 quantized versions. The model can be conveniently deployed on consumer GPUs like the Nvidia 3090 with almost no performance loss.
30
+ 4. **Open-source, free, and commercially usable**: Baichuan-13B is not only fully open to academic research, but developers can also use it for free commercially after applying for and receiving official commercial permission via email.
31
+
32
+
33
+ ## 使用方式
34
+
35
+ 如下是一个使用Baichuan-13B-Chat进行对话的示例，正确输出为""乔戈里峰。世界第二高峰———乔戈里峰西方登山者称其为k2峰，海拔高度是8611米，位于喀喇昆仑山脉的中巴边境上""
36
+ ```python
37
+ import torch
38
+ from transformers import AutoModelForCausalLM, AutoTokenizer
39
+ from transformers.generation.utils import GenerationConfig
40
+ tokenizer = AutoTokenizer.from_pretrained(""baichuan-inc/Baichuan-13B-Chat"", use_fast=False, trust_remote_code=True)
41
+ model = AutoModelForCausalLM.from_pretrained(""baichuan-inc/Baichuan-13B-Chat"", device_map=""auto"", torch_dtype=torch.float16, trust_remote_code=True)
42
+ model.generation_config = GenerationConfig.from_pretrained(""baichuan-inc/Baichuan-13B-Chat"")
43
+ messages = []
44
+ messages.append({""role"": ""user"", ""content"": ""世界上第二高的山峰是哪座""})
45
+ response = model.chat(tokenizer, messages)
46
+ print(response)
47
+ ```
48
+
49
+ Here is an example of a conversation using Baichuan-13B-Chat, the correct output is ""K2. The world's second highest peak - K2, also known as Mount Godwin-Austen or Chhogori, with an altitude of 8611 meters, is located on the China-Pakistan border in the Karakoram Range.""
50
+ ```python
51
+ import torch
52
+ from transformers import AutoModelForCausalLM, AutoTokenizer
53
+ from transformers.generation.utils import GenerationConfig
54
+ tokenizer = AutoTokenizer.from_pretrained(""baichuan-inc/Baichuan-13B-Chat"", use_fast=False, trust_remote_code=True)
55
+ model = AutoModelForCausalLM.from_pretrained(""baichuan-inc/Baichuan-13B-Chat"", device_map=""auto"", torch_dtype=torch.float16, trust_remote_code=True)
56
+ model.generation_config = GenerationConfig.from_pretrained(""baichuan-inc/Baichuan-13B-Chat"")
57
+ messages = []
58
+ messages.append({""role"": ""user"", ""content"": ""Which moutain is the second highest one in the world?""})
59
+ response = model.chat(tokenizer, messages)
60
+ print(response)
61
+ ```
62
+
63
+ ## 量化部署
64
+
65
+ Baichuan-13B 支持 int8 和 int4 量化，用户只需在推理代码中简单修改两行即可实现。请注意，如果是为了节省显存而进行量化，应加载原始精度模型到 CPU 后再开始量化；避免在 `from_pretrained` 时添加 `device_map='auto'` 或者其它会导致把原始精度模型直接加载到 GPU 的行为的参数。
66
+
67
+ Baichuan-13B supports int8 and int4 quantization, users only need to make a simple two-line change in the inference code to implement it. Please note, if quantization is done to save GPU memory, the original precision model should be loaded onto the CPU before starting quantization. Avoid adding parameters such as `device_map='auto'` or others that could cause the original precision model to be loaded directly onto the GPU when executing `from_pretrained`.
68
+
69
+ 使用 int8 量化 (To use int8 quantization):
70
+ ```python
71
+ model = AutoModelForCausalLM.from_pretrained(""baichuan-inc/Baichuan-13B-Chat"", torch_dtype=torch.float16, trust_remote_code=True)
72
+ model = model.quantize(8).cuda()
73
+ ```
74
+
75
+ 同样的，如需使用 int4 量化 (Similarly, to use int4 quantization):
76
+ ```python
77
+ model = AutoModelForCausalLM.from_pretrained(""baichuan-inc/Baichuan-13B-Chat"", torch_dtype=torch.float16, trust_remote_code=True)
78
+ model = model.quantize(4).cuda()
79
+ ```
80
+
81
+ ## 模型详情
82
+
83
+ ### 模型描述
84
+
85
+ <!-- Provide a longer summary of what this model is. -->
86
+
87
+ - **Developed by:** 百川智能(Baichuan Intelligent Technology)
88
+ - **Email**: opensource@baichuan-inc.com
89
+ - **Language(s) (NLP):** Chinese/English
90
+ - **License:** 【Community License for Baichuan-13B Model】([ZH](Baichuan-13B%20模型社区许可协议.pdf)|
91
+ [EN](Community%20License%20for%20Baichuan-13B%20Model.pdf))
92
+
93
+ **商业用途(For commercial use):** 请通过 [Email](mailto:opensource@baichuan-inc.com) 联系申请书面授权。(Contact us via [Email](mailto:opensource@baichuan-inc.com) above to apply for written authorization.)
94
+
95
+
96
+ ### 模型结构
97
+
98
+ <!-- Provide the basic links for the model. -->
99
+
100
+ 整体模型基于Baichuan-7B，为了获得更好的推理性能，Baichuan-13B 使用了 ALiBi 线性偏置技术，相对于 Rotary Embedding 计算量更小，对推理性能有显著提升；与标准的 LLaMA-13B 相比，生成 2000 个 tokens 的平均推理速度 (tokens/s)，实测提升 31.6%：
101
+
102
+ | Model | tokens/s |
103
+ |-------------|----------|
104
+ | LLaMA-13B | 19.4 |
105
+ | Baichuan-13B| 25.4 |
106
+
107
+ 具体参数和见下表
108
+ | 模型名称 | 隐含层维度 | 层数 | 头数 |词表大小 | 总参数量 | 训练数据（tokens） | 位置编码 | 最大长度 |
109
+ |-------------------------|-------|------------|------------|-----------------|--------|--------|----------------|---------|
110
+ | Baichuan-7B | 4,096 | 32 | 32 | 64,000 | 7,000,559,616 | 1.2万亿 | [RoPE](https://arxiv.org/abs/2104.09864) | 4,096 |
111
+ | Baichuan-13B | 5,120 | 40 | 40 | 64,000 | 13,264,901,120 | 1.4万亿 | [ALiBi](https://arxiv.org/abs/2108.12409) | 4,096
112
+
113
+ The overall model is based on Baichuan-7B. In order to achieve better inference performance, Baichuan-13B uses ALiBi linear bias technology, which has a smaller computational load compared to Rotary Embedding, and significantly improves inference performance. Compared with the standard LLaMA-13B, the average inference speed (tokens/s) for generating 2000 tokens has been tested to increase by 31.6%:
114
+
115
+ | Model | tokens/s |
116
+ |-------------|----------|
117
+ | LLaMA-13B | 19.4 |
118
+ | Baichuan-13B| 25.4 |
119
+
120
+ The specific parameters are as follows:
121
+ | Model Name | Hidden Size | Num Layers | Num Attention Heads |Vocab Size | Total Params | Training Dats（tokens） | Position Embedding | Max Length |
122
+ |-------------------------|-------|------------|------------|-----------------|--------|--------|----------------|---------|
123
+ | Baichuan-7B | 4,096 | 32 | 32 | 64,000 | 7,000,559,616 | 1.2万亿 | [RoPE](https://arxiv.org/abs/2104.09864) | 4,096 |
124
+ | Baichuan-13B | 5,120 | 40 | 40 | 64,000 | 13,264,901,120 | 1.4万亿 | [ALiBi](https://arxiv.org/abs/2108.12409) | 4,096
125
+
126
+ ## 使用须知
127
+
128
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
129
+
130
+
131
+ ### 免责声明
132
+
133
+ 我们在此声明，我们的开发团队并未基于 Baichuan-13B 模型开发任何应用，无论是在 iOS、Android、网页或任何其他平台。我们强烈呼吁所有使用者，不要利用 Baichuan-13B 模型进行任何危害国家社会安全或违法的活动。另外，我们也要求使用者不要将 Baichuan-13B 模型用于未经适当安全审查和备案的互联网服务。我们希望所有的使用者都能遵守这个原则，确保科技的发展能在规范和合法的环境下进行。
134
+
135
+ 我们已经尽我们所能，来确保模型训练过程中使用的数据的合规性。然而，尽管我们已经做出了巨大的努力，但由于模型和数据的复杂性，仍有可能存在一些无法预见的问题。因此，如果由于使用 Baichuan-13B 开源模型而导致的任何问题，包括但不限于数据安全问题、公共舆论风险，或模型被误导、滥用、传播或不当利用所带来的任何风险和问题，我们将不承担任何责任。
136
+
137
+ We hereby declare that our development team has not developed any applications based on the Baichuan-13B model, whether on iOS, Android, the web, or any other platform. We strongly urge all users not to use the Baichuan-13B model for any activities that harm national social security or are illegal. In addition, we also ask users not to use the Baichuan-13B model for internet services that have not undergone appropriate security review and filing. We hope that all users will adhere to this principle to ensure that technological development takes place in a regulated and legal environment.
138
+
139
+ We have done our utmost to ensure the compliance of the data used in the model training process. However, despite our great efforts, due to the complexity of the model and data, there may still be some unforeseen issues. Therefore, we will not take any responsibility for any issues arising from the use of the Baichuan-13B open-source model, including but not limited to data security issues, public opinion risks, or any risks and problems arising from the model being misled, misused, disseminated, or improperly exploited.
140
+
141
+ ## 训练详情
142
+
143
+ 训练具体设置参见[Baichuan-13B](https://github.com/baichuan-inc/Baichuan-13B)。
144
+
145
+ For specific training settings, please refer to [Baichuan-13B](https://github.com/baichuan-inc/Baichuan-13B).
146
+
147
+ ## 测评结果
148
+
149
+ ## [C-Eval](https://cevalbenchmark.com/index.html#home)
150
+
151
+ | Model 5-shot | STEM | Social Sciences | Humanities | Others | Average |
152
+ |-------------------------|:-----:|:---------------:|:----------:|:------:|:-------:|
153
+ | Baichuan-7B | 38.2 | 52.0 | 46.2 | 39.3 | 42.8 |
154
+ | Chinese-Alpaca-Plus-13B | 35.2 | 45.6 | 40.0 | 38.2 | 38.8 |
155
+ | Vicuna-13B | 30.5 | 38.2 | 32.5 | 32.5 | 32.8 |
156
+ | Chinese-LLaMA-Plus-13B | 30.3 | 38.0 | 32.9 | 29.1 | 32.1 |
157
+ | Ziya-LLaMA-13B-Pretrain | 27.6 | 34.4 | 32.0 | 28.6 | 30.0 |
158
+ | LLaMA-13B | 27.0 | 33.6 | 27.7 | 27.6 | 28.5 |
159
+ | moss-moon-003-base (16B)| 27.0 | 29.1 | 27.2 | 26.9 | 27.4 |
160
+ | **Baichuan-13B-Base** | **45.9** | **63.5** | **57.2** | **49.3** | **52.4** |
161
+ | **Baichuan-13B-Chat** | **43.7** | **64.6** | **56.2** | **49.2** | **51.5** |
162
+
163
+ ## [MMLU](https://arxiv.org/abs/2009.03300)
164
+
165
+ | Model 5-shot | STEM | Social Sciences | Humanities | Others | Average |
166
+ |-------------------------|:-----:|:---------------:|:----------:|:------:|:-------:|
167
+ | Vicuna-13B | 40.4 | 60.5 | 49.5 | 58.4 | 52.0 |
168
+ | LLaMA-13B | 36.1 | 53.0 | 44.0 | 52.8 | 46.3 |
169
+ | Chinese-Alpaca-Plus-13B | 36.9 | 48.9 | 40.5 | 50.5 | 43.9 |
170
+ | Ziya-LLaMA-13B-Pretrain | 35.6 | 47.6 | 40.1 | 49.4 | 42.9 |
171
+ | Baichuan-7B | 35.6 | 48.9 | 38.4 | 48.1 | 42.3 |
172
+ | Chinese-LLaMA-Plus-13B | 33.1 | 42.8 | 37.0 | 44.6 | 39.2 |
173
+ | moss-moon-003-base (16B)| 22.4 | 22.8 | 24.2 | 24.4 | 23.6 |
174
+ | **Baichuan-13B-Base** | **41.6** | **60.9** | **47.4** | **58.5** | **51.6** |
175
+ | **Baichuan-13B-Chat** | **40.9** | **60.9** | **48.8** | **59.0** | **52.1** |
176
+ > 说明：我们采用了 MMLU 官方的[评测方案](https://github.com/hendrycks/test)。
177
+
178
+ ## [CMMLU](https://github.com/haonan-li/CMMLU)
179
+
180
+ | Model 5-shot | STEM | Humanities | Social Sciences | Others | China Specific | Average |
181
+ |-------------------------|:-----:|:----------:|:---------------:|:------:|:--------------:|:-------:|
182
+ | Baichuan-7B | 34.4 | 47.5 | 47.6 | 46.6 | 44.3 | 44.0 |
183
+ | Vicuna-13B | 31.8 | 36.2 | 37.6 | 39.5 | 34.3 | 36.3 |
184
+ | Chinese-Alpaca-Plus-13B | 29.8 | 33.4 | 33.2 | 37.9 | 32.1 | 33.4 |
185
+ | Chinese-LLaMA-Plus-13B | 28.1 | 33.1 | 35.4 | 35.1 | 33.5 | 33.0 |
186
+ | Ziya-LLaMA-13B-Pretrain | 29.0 | 30.7 | 33.8 | 34.4 | 31.9 | 32.1 |
187
+ | LLaMA-13B | 29.2 | 30.8 | 31.6 | 33.0 | 30.5 | 31.2 |
188
+ | moss-moon-003-base (16B)| 27.2 | 30.4 | 28.8 | 32.6 | 28.7 | 29.6 |
189
+ | **Baichuan-13B-Base** | **41.7** | **61.1** | **59.8** | **59.0** | **56.4** | **55.3** |
190
+ | **Baichuan-13B-Chat** | **42.8** | **62.6** | **59.7** | **59.0** | **56.1** | **55.8** |
191
+ > 说明：CMMLU 是一个综合性的中文评估基准，专门用于评估语言模型在中文语境下的知识和推理能力。我们采用了其官方的[评测方案](https://github.com/haonan-li/CMMLU)。
192
+
193
+ ## 微信群组
194
+ ![WeChat](https://github.com/baichuan-inc/Baichuan-13B/blob/main/media/wechat.jpeg?raw=true)
195
+ ","{""id"": ""baichuan-inc/Baichuan-13B-Chat"", ""author"": ""baichuan-inc"", ""sha"": ""e3e1498bb6b7fb0bcb5a65679c09b862d7c29301"", ""last_modified"": ""2024-01-09 07:56:42+00:00"", ""created_at"": ""2023-07-08 05:58:27+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 3619, ""downloads_all_time"": null, ""likes"": 631, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""pytorch"", ""baichuan"", ""text-generation"", ""custom_code"", ""zh"", ""en"", ""arxiv:2104.09864"", ""arxiv:2108.12409"", ""arxiv:2009.03300"", ""autotrain_compatible"", ""text-generation-inference"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""language:\n- zh\n- en\npipeline_tag: text-generation\ninference: false"", ""widget_data"": [{""text"": ""\u6211\u53eb\u6731\u5229\u5b89\uff0c\u6211\u559c\u6b22""}, {""text"": ""\u6211\u53eb\u6258\u9a6c\u65af\uff0c\u6211\u7684\u4e3b\u8981""}, {""text"": ""\u6211\u53eb\u739b\u4e3d\u4e9a\uff0c\u6211\u6700\u559c\u6b22\u7684""}, {""text"": ""\u6211\u53eb\u514b\u62c9\u62c9\uff0c\u6211\u662f""}, {""text"": ""\u4ece\u524d\uff0c""}], ""model_index"": null, ""config"": {""architectures"": [""BaichuanForCausalLM""], ""auto_map"": {""AutoConfig"": ""configuration_baichuan.BaichuanConfig"", ""AutoModelForCausalLM"": ""modeling_baichuan.BaichuanForCausalLM""}, ""model_type"": ""baichuan"", ""tokenizer_config"": {""bos_token"": {""__type"": ""AddedToken"", ""content"": ""<s>"", ""lstrip"": false, ""normalized"": true, ""rstrip"": false, ""single_word"": true}, ""eos_token"": {""__type"": ""AddedToken"", ""content"": ""</s>"", ""lstrip"": false, ""normalized"": true, ""rstrip"": false, ""single_word"": true}, ""pad_token"": {""__type"": ""AddedToken"", ""content"": ""<unk>"", ""lstrip"": false, ""normalized"": true, ""rstrip"": false, ""single_word"": true}, ""unk_token"": {""__type"": ""AddedToken"", ""content"": ""<unk>"", ""lstrip"": false, ""normalized"": true, ""rstrip"": false, ""single_word"": true}}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": ""modeling_baichuan.BaichuanForCausalLM"", ""pipeline_tag"": ""text-generation"", ""processor"": null}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Baichuan-13B \u6a21\u578b\u793e\u533a\u8bb8\u53ef\u534f\u8bae.pdf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Community License for Baichuan-13B Model.pdf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='configuration_baichuan.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_utils.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='handler.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='modeling_baichuan.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00001-of-00003.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00002-of-00003.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00003-of-00003.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model.bin.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='quantizer.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='requirements.txt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenization_baichuan.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.model', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [""WinterGYC/BaiChuan-13B-Chat"", ""xuqinyang/Baichuan-13B-Chat"", ""EmbeddedLLM/chat-template-generation"", ""Justinrune/LLaMA-Factory"", ""kenken999/fastapi_django_main_live"", ""WinterGYC/Baichuan-13B-Chat-Int8"", ""Cran-May/yugangVI"", ""officialhimanshu595/llama-factory"", ""li-qing/FIRE"", ""WinterGYC/Baichuan-13B-Chat-Int8-Docker"", ""syx948/ChatPDF"", ""Zulelee/langchain-chatchat"", ""xuqinyang/Baichuan-13B-Chat-Int8-Cpp"", ""xuqinyang/Baichuan-13B-Chat-Int4-Cpp"", ""tianleliphoebe/visual-arena"", ""Cran-May/ygVI"", ""Ashmal/MobiLlama"", ""kaixuan42/aigc-lab"", ""WinterGYC/Baichuan-13B-Chat-Int8-Gradio"", ""PegaMichael/Taiwan-LLaMa2-Copy"", ""tjtanaa/chat-template-generation"", ""CaiRou-Huang/TwLLM7B-v2.0-base"", ""Havi999/FORAI"", ""pallavijaini/NeuralChat-LLAMA-POC"", ""blackwingedkite/gutalk"", ""cllatMTK/Breeze"", ""nengrenjie83/MedicalGPT-main"", ""shengzi/Baichuan-13B-Chat"", ""gordonchan/embedding-m3e-large"", ""larsthepenguin/trt-llm-rag-windows-main"", ""JiakunXu/Baichuan-13B-Chat"", ""blackwingedkite/alpaca2_clas"", ""Cran-May/ygVIB"", ""Cran-May/ygVIC"", ""jaekwon/intel_cpu_chat"", ""lianglv/NeuralChat-ICX-INT4"", ""hucoa/Baichuan-13B-Chat"", ""Bofeee5675/FIRE"", ""evelyn-lo/evelyn"", ""yuantao-infini-ai/demo_test"", ""zjasper666/bf16_vs_fp8"", ""martinakaduc/melt"", ""cloneQ/internLMRAG"", ""hujin0929/LlamaIndex_RAG"", ""flyfive0315/internLlamaIndex"", ""sunxiaokang/llamaindex_RAG_web"", ""kai119/llama"", ""qxy826982153/LlamaIndexRAG"", ""msun415/Llamole""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-01-09 07:56:42+00:00"", ""cardData"": ""language:\n- zh\n- en\npipeline_tag: text-generation\ninference: false"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": ""modeling_baichuan.BaichuanForCausalLM"", ""pipeline_tag"": ""text-generation"", ""processor"": null}, ""_id"": ""64a8fb03a8e86b27d097925b"", ""modelId"": ""baichuan-inc/Baichuan-13B-Chat"", ""usedStorage"": 79592869006}",0,,0,https://huggingface.co/shibing624/vicuna-baichuan-13b-chat-lora,1,,0,,0,"Ashmal/MobiLlama, Bofeee5675/FIRE, EmbeddedLLM/chat-template-generation, Justinrune/LLaMA-Factory, WinterGYC/BaiChuan-13B-Chat, Zulelee/langchain-chatchat, blackwingedkite/gutalk, evelyn-lo/evelyn, gordonchan/embedding-m3e-large, huggingface/InferenceSupport/discussions/new?title=baichuan-inc/Baichuan-13B-Chat&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bbaichuan-inc%2FBaichuan-13B-Chat%5D(%2Fbaichuan-inc%2FBaichuan-13B-Chat)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, kenken999/fastapi_django_main_live, martinakaduc/melt, xuqinyang/Baichuan-13B-Chat",13
Bio_ClinicalBERT_finetunes_20250426_121513.csv_finetunes_20250426_121513.csv ADDED
The diff for this file is too large to render. See raw diff
 
ChatTTS_finetunes_20250424_193500.csv_finetunes_20250424_193500.csv ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ 2Noise/ChatTTS,"---
3
+ license: cc-by-nc-4.0
4
+ library_name: chat_tts
5
+ pipeline_tag: text-to-audio
6
+ ---
7
+
8
+
9
+ **We are also training larger-scale models and need computational power and data support. If you can provide assistance, please contact OPEN-SOURCE@2NOISE.COM. Thank you very much.**
10
+
11
+ ## Clone the Repository
12
+ First, clone the Git repository:
13
+ ```bash
14
+ git clone https://github.com/2noise/ChatTTS.git
15
+ ```
16
+
17
+ ## Model Inference
18
+
19
+
20
+ ```python
21
+ # Import necessary libraries and configure settings
22
+ import torch
23
+ import torchaudio
24
+ torch._dynamo.config.cache_size_limit = 64
25
+ torch._dynamo.config.suppress_errors = True
26
+ torch.set_float32_matmul_precision('high')
27
+
28
+ import ChatTTS
29
+ from IPython.display import Audio
30
+
31
+ # Initialize and load the model:
32
+ chat = ChatTTS.Chat()
33
+ chat.load_models(compile=False) # Set to True for better performance
34
+
35
+ # Define the text input for inference (Support Batching)
36
+ texts = [
37
+ ""So we found being competitive and collaborative was a huge way of staying motivated towards our goals, so one person to call when you fall off, one person who gets you back on then one person to actually do the activity with."",
38
+ ]
39
+
40
+ # Perform inference and play the generated audio
41
+ wavs = chat.infer(texts)
42
+ Audio(wavs[0], rate=24_000, autoplay=True)
43
+
44
+ # Save the generated audio
45
+ torchaudio.save(""output.wav"", torch.from_numpy(wavs[0]), 24000)
46
+ ```
47
+ **For more usage examples, please refer to the [example notebook](https://github.com/2noise/ChatTTS/blob/main/example.ipynb), which includes parameters for finer control over the generated speech, such as specifying the speaker, adjusting speech speed, and adding laughter.**
48
+
49
+
50
+
51
+
52
+
53
+
54
+ ### Disclaimer: For Academic Purposes Only
55
+
56
+ The information provided in this document is for academic purposes only. It is intended for educational and research use, and should not be used for any commercial or legal purposes. The authors do not guarantee the accuracy, completeness, or reliability of the information.","{""id"": ""2Noise/ChatTTS"", ""author"": ""2Noise"", ""sha"": ""1a3c04a8b0651689bd9242fbb55b1f4b5a9aef84"", ""last_modified"": ""2024-10-22 08:26:20+00:00"", ""created_at"": ""2024-05-25 06:07:38+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 2918, ""downloads_all_time"": null, ""likes"": 1551, ""library_name"": ""chat_tts"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""chat_tts"", ""safetensors"", ""text-to-audio"", ""license:cc-by-nc-4.0"", ""region:us""], ""pipeline_tag"": ""text-to-audio"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""library_name: chat_tts\nlicense: cc-by-nc-4.0\npipeline_tag: text-to-audio"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/DVAE.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/DVAE.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/DVAE_full.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/Decoder.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/Decoder.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/Embed.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/GPT.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/Vocos.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/Vocos.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/gpt/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/gpt/model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/spk_stat.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/tokenizer.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/tokenizer/special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/tokenizer/tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/tokenizer/tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config/decoder.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config/dvae.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config/gpt.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config/path.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config/vocos.yaml', size=None, blob_id=None, lfs=None)""], ""spaces"": [""Dzkaka/ChatTTS"", ""lenML/ChatTTS-Forge"", ""Hilley/ChatTTS-OpenVoice"", ""markmagic/ChatTTS"", ""Hilley/ChatVC"", ""wffcyrus/ChatTTS-Story-Telling"", ""6Simple9/ChatTTS-OpenVoice"", ""fcyai/ChatTTS"", ""rao223/ChatTTS-Forge"", ""chenmgtea/chat-tts"", ""doby4u/chattts"", ""cbhhhcb/ChatTTS"", ""zelk12/ChatTTS-Forge_English_interface"", ""rainnee0925/ChatTTS"", ""prajjwalkapoor/tts"", ""fcyai/ChatTTS-Story-Telling"", ""sysf/ChatTTS"", ""savokiss/ChatTTS"", ""savokiss/chattts-free"", ""arpy8/chattts"", ""lisongfeng/ChatTTS-WebUI"", ""rainnee/ChatTTS"", ""wffcyrus/ChatTTS-Forge"", ""zzhouz/learningself"", ""docs4dev/gptalk"", ""sandy-try/ChatTTS-Forge"", ""slingkid/ChatVC2"", ""panyanyany/ChatTTS"", ""slingkid/ChatVC4"", ""Rdtuetr/ChatTTS"", ""hikerxu/ChatTTS"", ""zhzabcd/ChatTTS-Forge"", ""jdhsi/ChatTTS"", ""AwesomeK/ChatTTS-OpenVoice"", ""zhengr/ChatTTS-Forge"", ""zhengr/ChatTTS2"", ""weismart1807/Linly-Talker"", ""emilalvaro/ChatTTS-OpenVoice"", ""vuxuanhoan/ChatTTS-Forge"", ""MaktubCN/Chat-TTS"", ""EZMODEL/chattts-free"", ""chenjacky131/ChatTTS-Forge"", ""lalalic/chattts"", ""yamazing/ChatTTS"", ""zhangyanhua0913/ChatTTS-OpenVoice"", ""thierryguyot67/tts"", ""EagleW96/CC_ChatTTS_demo""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-10-22 08:26:20+00:00"", ""cardData"": ""library_name: chat_tts\nlicense: cc-by-nc-4.0\npipeline_tag: text-to-audio"", ""transformersInfo"": null, ""_id"": ""6651802a815d7642d5aaef56"", ""modelId"": ""2Noise/ChatTTS"", ""usedStorage"": 2364745599}",0,,0,,0,,0,,0,"6Simple9/ChatTTS-OpenVoice, Dzkaka/ChatTTS, EagleW96/CC_ChatTTS_demo, Hilley/ChatTTS-OpenVoice, Hilley/ChatVC, cbhhhcb/ChatTTS, huggingface/InferenceSupport/discussions/546, lenML/ChatTTS-Forge, panyanyany/ChatTTS, rao223/ChatTTS-Forge, savokiss/chattts-free, wffcyrus/ChatTTS-Forge, zelk12/ChatTTS-Forge_English_interface",13
ChilloutMix_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ AnonPerson/ChilloutMix,N/A,"{""id"": ""AnonPerson/ChilloutMix"", ""author"": ""AnonPerson"", ""sha"": ""65ee86993c78c48939d47719c6970c3315ca507d"", ""last_modified"": ""2023-02-22 14:04:42+00:00"", ""created_at"": ""2023-02-22 13:46:05+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 306, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": null, ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ChilloutMix-ni-fp16.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Japanese-doll-likeness.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Korean-doll-likeness.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Taiwan-doll-likeness.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ulzzang-6500.pt', size=None, blob_id=None, lfs=None)""], ""spaces"": [""INDONESIA-AI/Lobe"", ""Recahtrada/2nd2"", ""Rifd/Gxtaucok"", ""enochianborg/stable-diffusion-webui-vorstcavry"", ""ClipHamper/stable-diffusion-webui"", ""stable13/createimage7"", ""manivannan7gp/stable-diffusion-webui-vorstcavry"", ""henrysion/stable-diffusion-test"", ""vorstcavry/Anapnoe"", ""diego2554/stable-diffusion-SG"", ""vorstcavry/vorst-cavry-a1111-public"", ""Furinkaz/StableDiffusion""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-02-22 14:04:42+00:00"", ""cardData"": null, ""transformersInfo"": null, ""_id"": ""63f61c9d52799101f3cd4136"", ""modelId"": ""AnonPerson/ChilloutMix"", ""usedStorage"": 2585985291}",0,,0,,0,,0,,0,"ClipHamper/stable-diffusion-webui, Furinkaz/StableDiffusion, INDONESIA-AI/Lobe, Recahtrada/2nd2, Rifd/Gxtaucok, diego2554/stable-diffusion-SG, enochianborg/stable-diffusion-webui-vorstcavry, henrysion/stable-diffusion-test, huggingface/InferenceSupport/discussions/new?title=AnonPerson/ChilloutMix&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BAnonPerson%2FChilloutMix%5D(%2FAnonPerson%2FChilloutMix)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, manivannan7gp/stable-diffusion-webui-vorstcavry, stable13/createimage7, vorstcavry/Anapnoe, vorstcavry/vorst-cavry-a1111-public",13
CodeLlama-70b-hf_finetunes_20250426_121513.csv_finetunes_20250426_121513.csv ADDED
@@ -0,0 +1,145 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ codellama/CodeLlama-70b-hf,"---
3
+ language:
4
+ - code
5
+ pipeline_tag: text-generation
6
+ tags:
7
+ - llama-2
8
+ license: llama2
9
+ ---
10
+ # **Code Llama**
11
+ Code Llama is a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 70 billion parameters. This is the repository for the base 70B version in the Hugging Face Transformers format. This model is designed for general code synthesis and understanding. Links to other models can be found in the index at the bottom.
12
+
13
+ > [!NOTE]
14
+ > This is a non-official Code Llama repo. You can find the official Meta repository in the [Meta Llama organization](https://huggingface.co/meta-llama/CodeLlama-70b-hf).
15
+
16
+ | | Base Model | Python | Instruct |
17
+ | --- | ----------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- |
18
+ | 7B | [codellama/CodeLlama-7b-hf](https://huggingface.co/codellama/CodeLlama-7b-hf) | [codellama/CodeLlama-7b-Python-hf](https://huggingface.co/codellama/CodeLlama-7b-Python-hf) | [codellama/CodeLlama-7b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-7b-Instruct-hf) |
19
+ | 13B | [codellama/CodeLlama-13b-hf](https://huggingface.co/codellama/CodeLlama-13b-hf) | [codellama/CodeLlama-13b-Python-hf](https://huggingface.co/codellama/CodeLlama-13b-Python-hf) | [codellama/CodeLlama-13b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-13b-Instruct-hf) |
20
+ | 34B | [codellama/CodeLlama-34b-hf](https://huggingface.co/codellama/CodeLlama-34b-hf) | [codellama/CodeLlama-34b-Python-hf](https://huggingface.co/codellama/CodeLlama-34b-Python-hf) | [codellama/CodeLlama-34b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-34b-Instruct-hf) |
21
+ | 70B | [codellama/CodeLlama-70b-hf](https://huggingface.co/codellama/CodeLlama-70b-hf) | [codellama/CodeLlama-70b-Python-hf](https://huggingface.co/codellama/CodeLlama-70b-Python-hf) | [codellama/CodeLlama-70b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-70b-Instruct-hf) |
22
+
23
+ ## Model Use
24
+
25
+ To use this model, please make sure to install `transformers`.
26
+
27
+ ```bash
28
+ pip install transformers accelerate
29
+ ```
30
+
31
+ Model capabilities:
32
+
33
+ - [x] Code completion.
34
+ - [ ] Infilling.
35
+ - [ ] Instructions / chat.
36
+ - [ ] Python specialist.
37
+
38
+ ## Model Details
39
+ *Note: Use of this model is governed by the Meta license. Meta developed and publicly released the Code Llama family of large language models (LLMs).
40
+
41
+ **Model Developers** Meta
42
+
43
+ **Variations** Code Llama comes in four model sizes, and three variants:
44
+
45
+ * Code Llama: base models designed for general code synthesis and understanding
46
+ * Code Llama - Python: designed specifically for Python
47
+ * Code Llama - Instruct: for instruction following and safer deployment
48
+
49
+ All variants are available in sizes of 7B, 13B, 34B, and 70B parameters.
50
+
51
+ **This repository contains the base version of the 70B parameters model.**
52
+
53
+ **Input** Models input text only.
54
+
55
+ **Output** Models generate text only.
56
+
57
+ **Model Architecture** Code Llama is an auto-regressive language model that uses an optimized transformer architecture. It was fine-tuned with up to 16k tokens and supports up to 100k tokens at inference time.
58
+
59
+ **Model Dates** Code Llama and its variants have been trained between January 2023 and January 2024.
60
+
61
+ **Status** This is a static model trained on an offline dataset. Future versions of Code Llama - Instruct will be released as we improve model safety with community feedback.
62
+
63
+ **License** A custom commercial license is available at: [https://ai.meta.com/resources/models-and-libraries/llama-downloads/](https://ai.meta.com/resources/models-and-libraries/llama-downloads/)
64
+
65
+ **Research Paper** More information can be found in the paper ""[Code Llama: Open Foundation Models for Code](https://ai.meta.com/research/publications/code-llama-open-foundation-models-for-code/)"" or its [arXiv page](https://arxiv.org/abs/2308.12950).
66
+
67
+ ## Intended Use
68
+ **Intended Use Cases** Code Llama and its variants are intended for commercial and research use in English and relevant programming languages. The base model Code Llama can be adapted for a variety of code synthesis and understanding tasks, Code Llama - Python is designed specifically to handle the Python programming language, and Code Llama - Instruct is intended to be safer to use for code assistant and generation applications.
69
+
70
+ **Out-of-Scope Uses** Use in any manner that violates applicable laws or regulations (including trade compliance laws). Use in languages other than English. Use in any other way that is prohibited by the Acceptable Use Policy and Licensing Agreement for Code Llama and its variants.
71
+
72
+ ## Hardware and Software
73
+ **Training Factors** We used custom training libraries. The training and fine-tuning of the released models have been performed Meta’s Research Super Cluster.
74
+
75
+ **Carbon Footprint** In aggregate, training all 12 Code Llama models required 1400K GPU hours of computation on hardware of type A100-80GB (TDP of 350-400W). Estimated total emissions were 228.55 tCO2eq, 100% of which were offset by Meta’s sustainability program.
76
+
77
+
78
+ ## Evaluation Results
79
+
80
+ See evaluations for the main models and detailed ablations in Section 3 and safety evaluations in Section 4 of the research paper.
81
+
82
+ ## Ethical Considerations and Limitations
83
+
84
+ Code Llama and its variants are a new technology that carries risks with use. Testing conducted to date has been in English, and has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, Code Llama’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate or objectionable responses to user prompts. Therefore, before deploying any applications of Code Llama, developers should perform safety testing and tuning tailored to their specific applications of the model.
85
+
86
+ Please see the Responsible Use Guide available available at [https://ai.meta.com/llama/responsible-use-guide](https://ai.meta.com/llama/responsible-use-guide).
87
+ ","{""id"": ""codellama/CodeLlama-70b-hf"", ""author"": ""codellama"", ""sha"": ""cc11b0f79c8a072bd1e1f32ec280e1ff8ec018f6"", ""last_modified"": ""2024-04-12 14:17:44+00:00"", ""created_at"": ""2024-01-29 10:59:03+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 378, ""downloads_all_time"": null, ""likes"": 314, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""pytorch"", ""safetensors"", ""llama"", ""text-generation"", ""llama-2"", ""code"", ""arxiv:2308.12950"", ""license:llama2"", ""autotrain_compatible"", ""text-generation-inference"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""language:\n- code\nlicense: llama2\npipeline_tag: text-generation\ntags:\n- llama-2"", ""widget_data"": null, ""model_index"": null, ""config"": {""architectures"": [""LlamaForCausalLM""], ""model_type"": ""llama"", ""tokenizer_config"": {""bos_token"": ""<s>"", ""eos_token"": ""</s>"", ""pad_token"": null, ""unk_token"": ""<unk>"", ""use_default_system_prompt"": false}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00001-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00002-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00003-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00004-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00005-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00006-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00007-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00008-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00009-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00010-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00011-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00012-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00013-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00014-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00015-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00016-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00017-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00018-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00019-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00020-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00021-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00022-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00023-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00024-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00025-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00026-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00027-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00028-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00029-of-00029.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00001-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00002-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00003-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00004-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00005-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00006-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00007-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00008-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00009-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00010-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00011-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00012-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00013-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00014-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00015-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00016-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00017-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00018-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00019-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00020-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00021-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00022-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00023-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00024-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00025-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00026-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00027-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00028-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00029-of-00029.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model.bin.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.model', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [""bigcode/bigcode-models-leaderboard"", ""KBaba7/Quant"", ""meval/multilingual-chatbot-arena-leaderboard"", ""prometheus-eval/BiGGen-Bench-Leaderboard"", ""HPAI-BSC/TuRTLe-Leaderboard"", ""bhaskartripathi/LLM_Quantization"", ""totolook/Quant"", ""FallnAI/Quantize-HF-Models"", ""Generatia/codellama-CodeLlama-70b-hf"", ""Alejo2639/codellama-CodeLlama-70b-hf"", ""ruslanmv/convert_to_gguf"", ""Irishcoder/codellama-CodeLlama-70b-hf"", ""Dineth1222/code_nova"", ""MetaReps/codellama-CodeLlama-70b-hf"", ""shrimantasatpati/codellama-CodeLlama-70b-hf"", ""WSLX/codellama-CodeLlama-70b-hf"", ""davila7/codellama-CodeLlama-70b-hf"", ""parthu10/codellama-CodeLlama-70b-hf"", ""HunterThief/codellama-CodeLlama-70b-hf"", ""jwebber/codellama-CodeLlama-70b-hf"", ""zanjani1/codellama-CodeLlama-70b-hf"", ""ozzy1987/codellama-CodeLlama-70b-hf"", ""taco1/codellama-CodeLlama-70b-hf"", ""voidnullnil/codellama-CodeLlama-70b-hf"", ""powolnik/codellama-CodeLlama-70b-hf"", ""MadK/codellama-CodeLlama-70b-hf"", ""An-Egoistic-Developer-Full-Of-Knowledge/codellama-CodeLlama-70b-hf"", ""An-Egoistic-Developer-Full-Of-Knowledge/codellama-CodeLlama-70b-hf1"", ""mucahitkayadan/codellama-CodeLlama-70b-hf"", ""ahsabbir104/codellama-CodeLlama-70b-hf"", ""dbasu/multilingual-chatbot-arena-leaderboard"", ""atlasas/bigcode-models-leaderboard"", ""Dugelon/First_agent_template"", ""K00B404/LLM_Quantization""], ""safetensors"": {""parameters"": {""BF16"": 68976910336}, ""total"": 68976910336}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-04-12 14:17:44+00:00"", ""cardData"": ""language:\n- code\nlicense: llama2\npipeline_tag: text-generation\ntags:\n- llama-2"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""_id"": ""65b784f7e886c5d4fdb1e02b"", ""modelId"": ""codellama/CodeLlama-70b-hf"", ""usedStorage"": 276039624093}",0,https://huggingface.co/nisten/BigCodeLlama-169b,1,https://huggingface.co/wisdominanutshell/splitter_70b_70B,1,"https://huggingface.co/TheBloke/CodeLlama-70B-hf-GGUF, https://huggingface.co/TheBloke/CodeLlama-70B-hf-AWQ, https://huggingface.co/TheBloke/CodeLlama-70B-hf-GPTQ, https://huggingface.co/mlc-ai/CodeLlama-70b-hf-q3f16_1-MLC, https://huggingface.co/mlc-ai/CodeLlama-70b-hf-q4f16_1-MLC, https://huggingface.co/mlc-ai/CodeLlama-70b-hf-q4f32_1-MLC, https://huggingface.co/mradermacher/CodeLlama-70b-hf-GGUF, https://huggingface.co/mradermacher/CodeLlama-70b-hf-i1-GGUF, https://huggingface.co/tensorblock/CodeLlama-70b-hf-GGUF",9,https://huggingface.co/Blazgo/2-coder-pro,1,"FallnAI/Quantize-HF-Models, HPAI-BSC/TuRTLe-Leaderboard, K00B404/LLM_Quantization, KBaba7/Quant, atlasas/bigcode-models-leaderboard, bhaskartripathi/LLM_Quantization, bigcode/bigcode-models-leaderboard, dbasu/multilingual-chatbot-arena-leaderboard, huggingface/InferenceSupport/discussions/new?title=codellama/CodeLlama-70b-hf&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bcodellama%2FCodeLlama-70b-hf%5D(%2Fcodellama%2FCodeLlama-70b-hf)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, meval/multilingual-chatbot-arena-leaderboard, parthu10/codellama-CodeLlama-70b-hf, prometheus-eval/BiGGen-Bench-Leaderboard, zanjani1/codellama-CodeLlama-70b-hf",13
88
+ nisten/BigCodeLlama-169b,"---
89
+ base_model: [codellama/CodeLlama-70b-hf]
90
+ tags:
91
+ - mergekit
92
+ - merge
93
+ - code
94
+ license: mit
95
+ pipeline_tag: conversational
96
+ ---
97
+ # BigCodeLLama LFG 🚀
98
+
99
+ ## Experimental CodeLlaMA frankenstein to see how it benchmarks
100
+
101
+ ### Models Merged with base ```codellama/CodeLlama-70b-hf```
102
+
103
+
104
+
105
+ The following models were included in the merge:
106
+ * ../CodeLlama-70b-hf
107
+ * ../CodeLlama-70b-Instruct-hf
108
+ * ../CodeLlama-70b-Python-hf
109
+
110
+ ### Configuration
111
+
112
+ The following YAML configuration was used to produce this model:
113
+
114
+ ```yaml
115
+ dtype: bfloat16
116
+ merge_method: passthrough
117
+ slices:
118
+ - sources:
119
+ - layer_range: [0, 69]
120
+ model:
121
+ model:
122
+ path: ../CodeLlama-70b-hf
123
+ - sources:
124
+ - layer_range: [66, 76]
125
+ model:
126
+ model:
127
+ path: ../CodeLlama-70b-Instruct-hf
128
+ - sources:
129
+ - layer_range: [42, 66]
130
+ model:
131
+ model:
132
+ path: ../CodeLlama-70b-hf
133
+ - sources:
134
+ - layer_range: [13, 37]
135
+ model:
136
+ model:
137
+ path: ../CodeLlama-70b-Python-hf
138
+ - sources:
139
+ - layer_range: [10, 80]
140
+ model:
141
+ model:
142
+ path: ../CodeLlama-70b-Instruct-hf
143
+ ```
144
+
145
+ ### Stay tuned for GGUFs quants","{""id"": ""nisten/BigCodeLlama-169b"", ""author"": ""nisten"", ""sha"": ""3b3e55cffbc6c2836f10183bd00d814d14e2fd43"", ""last_modified"": ""2024-01-30 02:11:18+00:00"", ""created_at"": ""2024-01-29 20:26:40+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 2, ""downloads_all_time"": null, ""likes"": 14, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""safetensors"", ""llama"", ""text-generation"", ""mergekit"", ""merge"", ""code"", ""conversational"", ""base_model:codellama/CodeLlama-70b-hf"", ""base_model:finetune:codellama/CodeLlama-70b-hf"", ""license:mit"", ""autotrain_compatible"", ""text-generation-inference"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- codellama/CodeLlama-70b-hf\nlicense: mit\npipeline_tag: conversational\ntags:\n- mergekit\n- merge\n- code"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": null, ""config"": {""architectures"": [""LlamaForCausalLM""], ""model_type"": ""llama"", ""tokenizer_config"": {""bos_token"": ""<s>"", ""eos_token"": ""</s>"", ""pad_token"": null, ""unk_token"": ""<unk>"", ""use_default_system_prompt"": false}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='mergekit_config.yml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00001-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00002-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00003-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00004-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00005-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00006-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00007-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00008-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00009-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00010-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00011-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00012-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00013-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00014-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00015-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00016-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00017-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00018-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00019-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00020-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00021-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00022-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00023-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00024-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00025-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00026-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00027-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00028-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00029-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00030-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00031-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00032-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00033-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00034-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00035-of-00035.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.model', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [""powolnik/nisten-BigCodeLlama-169b""], ""safetensors"": {""parameters"": {""BF16"": 169088475136}, ""total"": 169088475136}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-01-30 02:11:18+00:00"", ""cardData"": ""base_model:\n- codellama/CodeLlama-70b-hf\nlicense: mit\npipeline_tag: conversational\ntags:\n- mergekit\n- merge\n- code"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""_id"": ""65b80a00db07f4cdc46861af"", ""modelId"": ""nisten/BigCodeLlama-169b"", ""usedStorage"": 338177659137}",1,,0,,0,,0,,0,"huggingface/InferenceSupport/discussions/new?title=nisten/BigCodeLlama-169b&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bnisten%2FBigCodeLlama-169b%5D(%2Fnisten%2FBigCodeLlama-169b)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, powolnik/nisten-BigCodeLlama-169b",2
DeBERTa-v3-base-mnli-fever-anli_finetunes_20250426_215237.csv_finetunes_20250426_215237.csv ADDED
The diff for this file is too large to render. See raw diff
 
DeepSeek-R1-GGUF_finetunes_20250424_223250.csv_finetunes_20250424_223250.csv ADDED
@@ -0,0 +1,442 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ unsloth/DeepSeek-R1-GGUF,"---
3
+ base_model: deepseek-ai/DeepSeek-R1
4
+ language:
5
+ - en
6
+ library_name: transformers
7
+ license: mit
8
+ tags:
9
+ - deepseek
10
+ - unsloth
11
+ - transformers
12
+ ---
13
+
14
+ <div>
15
+ <p style=""margin-bottom: 0; margin-top: 0;"">
16
+ <strong>See <a href=""https://huggingface.co/collections/unsloth/deepseek-r1-all-versions-678e1c48f5d2fce87892ace5"">our collection</a> for versions of Deepseek-R1 including GGUF & 4-bit formats.</strong>
17
+ </p>
18
+ <p style=""margin-bottom: 0;"">
19
+ <em>Unsloth's DeepSeek-R1 <a href=""https://unsloth.ai/blog/deepseekr1-dynamic"">1.58-bit + 2-bit Dynamic Quants</a> is selectively quantized, greatly improving accuracy over standard 1-bit/2-bit.</em>
20
+ </p>
21
+ <div style=""display: flex; gap: 5px; align-items: center; "">
22
+ <a href=""https://github.com/unslothai/unsloth/"">
23
+ <img src=""https://github.com/unslothai/unsloth/raw/main/images/unsloth%20new%20logo.png"" width=""133"">
24
+ </a>
25
+ <a href=""https://discord.gg/unsloth"">
26
+ <img src=""https://github.com/unslothai/unsloth/raw/main/images/Discord%20button.png"" width=""173"">
27
+ </a>
28
+ <a href=""https://docs.unsloth.ai/basics/tutorial-how-to-run-deepseek-r1-on-your-own-local-device"">
29
+ <img src=""https://raw.githubusercontent.com/unslothai/unsloth/refs/heads/main/images/documentation%20green%20button.png"" width=""143"">
30
+ </a>
31
+ </div>
32
+ <h1 style=""margin-top: 0rem;"">Instructions to run this model in llama.cpp:</h2>
33
+ </div>
34
+
35
+ Or you can view more detailed instructions here: [unsloth.ai/blog/deepseekr1-dynamic](https://unsloth.ai/blog/deepseekr1-dynamic)
36
+ 1. Do not forget about `<｜User｜>` and `<｜Assistant｜>` tokens! - Or use a chat template formatter
37
+ 2. Obtain the latest `llama.cpp` at https://github.com/ggerganov/llama.cpp. You can follow the build instructions below as well:
38
+ ```bash
39
+ apt-get update
40
+ apt-get install build-essential cmake curl libcurl4-openssl-dev -y
41
+ git clone https://github.com/ggerganov/llama.cpp
42
+ cmake llama.cpp -B llama.cpp/build \
43
+ -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=ON -DLLAMA_CURL=ON
44
+ cmake --build llama.cpp/build --config Release -j --clean-first --target llama-quantize llama-cli llama-gguf-split
45
+ cp llama.cpp/build/bin/llama-* llama.cpp
46
+ ```
47
+ 3. It's best to use `--min-p 0.05` to counteract very rare token predictions - I found this to work well especially for the 1.58bit model.
48
+ 4. Download the model via:
49
+ ```python
50
+ # pip install huggingface_hub hf_transfer
51
+ # import os # Optional for faster downloading
52
+ # os.environ[""HF_HUB_ENABLE_HF_TRANSFER""] = ""1""
53
+
54
+ from huggingface_hub import snapshot_download
55
+ snapshot_download(
56
+ repo_id = ""unsloth/DeepSeek-R1-GGUF"",
57
+ local_dir = ""DeepSeek-R1-GGUF"",
58
+ allow_patterns = [""*UD-IQ1_S*""], # Select quant type UD-IQ1_S for 1.58bit
59
+ )
60
+ ```
61
+ 5. Example with Q4_0 K quantized cache **Notice -no-cnv disables auto conversation mode**
62
+ ```bash
63
+ ./llama.cpp/llama-cli \
64
+ --model DeepSeek-R1-GGUF/DeepSeek-R1-UD-IQ1_S/DeepSeek-R1-UD-IQ1_S-00001-of-00003.gguf \
65
+ --cache-type-k q4_0 \
66
+ --threads 12 -no-cnv --prio 2 \
67
+ --temp 0.6 \
68
+ --ctx-size 8192 \
69
+ --seed 3407 \
70
+ --prompt ""<｜User｜>Create a Flappy Bird game in Python.<｜Assistant｜>""
71
+ ```
72
+ Example output:
73
+
74
+ ```txt
75
+ <think>
76
+ Okay, so I need to figure out what 1 plus 1 is. Hmm, where do I even start? I remember from school that adding numbers is pretty basic, but I want to make sure I understand it properly.
77
+ Let me think, 1 plus 1. So, I have one item and I add another one. Maybe like a apple plus another apple. If I have one apple and someone gives me another, I now have two apples. So, 1 plus 1 should be 2. That makes sense.
78
+ Wait, but sometimes math can be tricky. Could it be something else? Like, in a different number system maybe? But I think the question is straightforward, using regular numbers, not like binary or hexadecimal or anything.
79
+ I also recall that in arithmetic, addition is combining quantities. So, if you have two quantities of 1, combining them gives you a total of 2. Yeah, that seems right.
80
+ Is there a scenario where 1 plus 1 wouldn't be 2? I can't think of any...
81
+ ```
82
+
83
+ 6. If you have a GPU (RTX 4090 for example) with 24GB, you can offload multiple layers to the GPU for faster processing. If you have multiple GPUs, you can probably offload more layers.
84
+ ```bash
85
+ ./llama.cpp/llama-cli \
86
+ --model DeepSeek-R1-GGUF/DeepSeek-R1-UD-IQ1_S/DeepSeek-R1-UD-IQ1_S-00001-of-00003.gguf \
87
+ --cache-type-k q4_0 \
88
+ --threads 12 -no-cnv --prio 2 \
89
+ --n-gpu-layers 7 \
90
+ --temp 0.6 \
91
+ --ctx-size 8192 \
92
+ --seed 3407 \
93
+ --prompt ""<｜User｜>Create a Flappy Bird game in Python.<｜Assistant｜>""
94
+ ```
95
+ 7. If you want to merge the weights together, use this script:
96
+ ```
97
+ ./llama.cpp/llama-gguf-split --merge \
98
+ DeepSeek-R1-GGUF/DeepSeek-R1-UD-IQ1_S-00001-of-00003.gguf \
99
+ merged_file.gguf
100
+ ```
101
+
102
+ | MoE Bits | Type | Disk Size | Accuracy | Link | Details |
103
+ | -------- | -------- | ------------ | ------------ | ---------------------| ---------- |
104
+ | 1.58bit | UD-IQ1_S | **131GB** | Fair | [Link](https://huggingface.co/unsloth/DeepSeek-R1-GGUF/tree/main/DeepSeek-R1-UD-IQ1_S) | MoE all 1.56bit. `down_proj` in MoE mixture of 2.06/1.56bit |
105
+ | 1.73bit | UD-IQ1_M | **158GB** | Good | [Link](https://huggingface.co/unsloth/DeepSeek-R1-GGUF/tree/main/DeepSeek-R1-UD-IQ1_M) | MoE all 1.56bit. `down_proj` in MoE left at 2.06bit |
106
+ | 2.22bit | UD-IQ2_XXS | **183GB** | Better | [Link](https://huggingface.co/unsloth/DeepSeek-R1-GGUF/tree/main/DeepSeek-R1-UD-IQ2_XXS) | MoE all 2.06bit. `down_proj` in MoE mixture of 2.5/2.06bit |
107
+ | 2.51bit | UD-Q2_K_XL | **212GB** | Best | [Link](https://huggingface.co/unsloth/DeepSeek-R1-GGUF/tree/main/DeepSeek-R1-UD-Q2_K_XL) | MoE all 2.5bit. `down_proj` in MoE mixture of 3.5/2.5bit |
108
+
109
+ # Finetune your own Reasoning model like R1 with Unsloth!
110
+ We have a free Google Colab notebook for turning Llama 3.1 (8B) into a reasoning model: https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Llama3.1_(8B)-GRPO.ipynb
111
+
112
+ [<img src=""https://raw.githubusercontent.com/unslothai/unsloth/main/images/Discord%20button.png"" width=""200""/>](https://discord.gg/unsloth)
113
+ [<img src=""https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png"" width=""200""/>](https://github.com/unslothai/unsloth)
114
+
115
+
116
+ ## ✨ Finetune for Free
117
+
118
+ All notebooks are **beginner friendly**! Add your dataset, click ""Run All"", and you'll get a 2x faster finetuned model which can be exported to GGUF, vLLM or uploaded to Hugging Face.
119
+
120
+ | Unsloth supports | Free Notebooks | Performance | Memory use |
121
+ |-----------------|--------------------------------------------------------------------------------------------------------------------------|-------------|----------|
122
+ | **GRPO with Phi-4 (14B)** | [▶️ Start on Colab](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Phi_4_(14B)-GRPO.ipynb) | 2x faster | 80% less |
123
+ | **Llama-3.2 (3B)** | [▶️ Start on Colab](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Llama3.2_(1B_and_3B)-Conversational.ipynb) | 2.4x faster | 58% less |
124
+ | **Llama-3.2 (11B vision)** | [▶️ Start on Colab](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Llama3.2_(11B)-Vision.ipynb) | 2x faster | 60% less |
125
+ | **Qwen2 VL (7B)** | [▶️ Start on Colab](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Qwen2_VL_(7B)-Vision.ipynb) | 1.8x faster | 60% less |
126
+ | **Qwen2.5 (7B)** | [▶️ Start on Colab](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Qwen2.5_(7B)-Alpaca.ipynb) | 2x faster | 60% less |
127
+ | **Llama-3.1 (8B)** | [▶️ Start on Colab](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Llama3.1_(8B)-Alpaca.ipynb) | 2.4x faster | 58% less |
128
+ | **Phi-3.5 (mini)** | [▶️ Start on Colab](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Phi_3.5_Mini-Conversational.ipynb) | 2x faster | 50% less |
129
+ | **Gemma 2 (9B)** | [▶️ Start on Colab](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Gemma2_(9B)-Alpaca.ipynb) | 2.4x faster | 58% less |
130
+ | **Mistral (7B)** | [▶️ Start on Colab](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Mistral_v0.3_(7B)-Conversational.ipynb) | 2.2x faster | 62% less |
131
+
132
+ [<img src=""https://raw.githubusercontent.com/unslothai/unsloth/refs/heads/main/images/documentation%20green%20button.png"" width=""200""/>](https://docs.unsloth.ai)
133
+
134
+ - This [Llama 3.2 conversational notebook](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Llama3.2_(1B_and_3B)-Conversational.ipynb) is useful for ShareGPT ChatML / Vicuna templates.
135
+ - This [text completion notebook](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Mistral_(7B)-Text_Completion.ipynb) is for raw text. This [DPO notebook](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) replicates Zephyr.
136
+ - \* Kaggle has 2x T4s, but we use 1. Due to overhead, 1x T4 is 5x faster.
137
+
138
+ ## Special Thanks
139
+ A huge thank you to the DeepSeek team for creating and releasing these models.
140
+
141
+
142
+
143
+ # DeepSeek-R1
144
+ <!-- markdownlint-disable first-line-h1 -->
145
+ <!-- markdownlint-disable html -->
146
+ <!-- markdownlint-disable no-duplicate-header -->
147
+
148
+ <div align=""center"">
149
+ <img src=""https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg?raw=true"" width=""60%"" alt=""DeepSeek-V3"" />
150
+ </div>
151
+ <hr>
152
+ <div align=""center"" style=""line-height: 1;"">
153
+ <a href=""https://www.deepseek.com/"" target=""_blank"" style=""margin: 2px;"">
154
+ <img alt=""Homepage"" src=""https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/badge.svg?raw=true"" style=""display: inline-block; vertical-align: middle;""/>
155
+ </a>
156
+ <a href=""https://chat.deepseek.com/"" target=""_blank"" style=""margin: 2px;"">
157
+ <img alt=""Chat"" src=""https://img.shields.io/badge/🤖%20Chat-DeepSeek%20R1-536af5?color=536af5&logoColor=white"" style=""display: inline-block; vertical-align: middle;""/>
158
+ </a>
159
+ <a href=""https://huggingface.co/deepseek-ai"" target=""_blank"" style=""margin: 2px;"">
160
+ <img alt=""Hugging Face"" src=""https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-DeepSeek%20AI-ffc107?color=ffc107&logoColor=white"" style=""display: inline-block; vertical-align: middle;""/>
161
+ </a>
162
+ </div>
163
+
164
+ <div align=""center"" style=""line-height: 1;"">
165
+ <a href=""https://discord.gg/Tc7c45Zzu5"" target=""_blank"" style=""margin: 2px;"">
166
+ <img alt=""Discord"" src=""https://img.shields.io/badge/Discord-DeepSeek%20AI-7289da?logo=discord&logoColor=white&color=7289da"" style=""display: inline-block; vertical-align: middle;""/>
167
+ </a>
168
+ <a href=""https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/qr.jpeg?raw=true"" target=""_blank"" style=""margin: 2px;"">
169
+ <img alt=""Wechat"" src=""https://img.shields.io/badge/WeChat-DeepSeek%20AI-brightgreen?logo=wechat&logoColor=white"" style=""display: inline-block; vertical-align: middle;""/>
170
+ </a>
171
+ <a href=""https://twitter.com/deepseek_ai"" target=""_blank"" style=""margin: 2px;"">
172
+ <img alt=""Twitter Follow"" src=""https://img.shields.io/badge/Twitter-deepseek_ai-white?logo=x&logoColor=white"" style=""display: inline-block; vertical-align: middle;""/>
173
+ </a>
174
+ </div>
175
+
176
+ <div align=""center"" style=""line-height: 1;"">
177
+ <a href=""https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-CODE"" style=""margin: 2px;"">
178
+ <img alt=""Code License"" src=""https://img.shields.io/badge/Code_License-MIT-f5de53?&color=f5de53"" style=""display: inline-block; vertical-align: middle;""/>
179
+ </a>
180
+ <a href=""https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE-MODEL"" style=""margin: 2px;"">
181
+ <img alt=""Model License"" src=""https://img.shields.io/badge/Model_License-Model_Agreement-f5de53?&color=f5de53"" style=""display: inline-block; vertical-align: middle;""/>
182
+ </a>
183
+ </div>
184
+
185
+
186
+ <p align=""center"">
187
+ <a href=""https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf""><b>Paper Link</b>👁️</a>
188
+ </p>
189
+
190
+
191
+ ## 1. Introduction
192
+
193
+ We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1.
194
+ DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning.
195
+ With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors.
196
+ However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing. To address these issues and further enhance reasoning performance,
197
+ we introduce DeepSeek-R1, which incorporates cold-start data before RL.
198
+ DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks.
199
+ To support the research community, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six dense models distilled from DeepSeek-R1 based on Llama and Qwen. DeepSeek-R1-Distill-Qwen-32B outperforms OpenAI-o1-mini across various benchmarks, achieving new state-of-the-art results for dense models.
200
+
201
+ **NOTE: Before running DeepSeek-R1 series models locally, we kindly recommend reviewing the [Usage Recommendation](#usage-recommendations) section.**
202
+
203
+ <p align=""center"">
204
+ <img width=""80%"" src=""figures/benchmark.jpg"">
205
+ </p>
206
+
207
+ ## 2. Model Summary
208
+
209
+ ---
210
+
211
+ **Post-Training: Large-Scale Reinforcement Learning on the Base Model**
212
+
213
+ - We directly apply reinforcement learning (RL) to the base model without relying on supervised fine-tuning (SFT) as a preliminary step. This approach allows the model to explore chain-of-thought (CoT) for solving complex problems, resulting in the development of DeepSeek-R1-Zero. DeepSeek-R1-Zero demonstrates capabilities such as self-verification, reflection, and generating long CoTs, marking a significant milestone for the research community. Notably, it is the first open research to validate that reasoning capabilities of LLMs can be incentivized purely through RL, without the need for SFT. This breakthrough paves the way for future advancements in this area.
214
+
215
+ - We introduce our pipeline to develop DeepSeek-R1. The pipeline incorporates two RL stages aimed at discovering improved reasoning patterns and aligning with human preferences, as well as two SFT stages that serve as the seed for the model's reasoning and non-reasoning capabilities.
216
+ We believe the pipeline will benefit the industry by creating better models.
217
+
218
+ ---
219
+
220
+ **Distillation: Smaller Models Can Be Powerful Too**
221
+
222
+ - We demonstrate that the reasoning patterns of larger models can be distilled into smaller models, resulting in better performance compared to the reasoning patterns discovered through RL on small models. The open source DeepSeek-R1, as well as its API, will benefit the research community to distill better smaller models in the future.
223
+ - Using the reasoning data generated by DeepSeek-R1, we fine-tuned several dense models that are widely used in the research community. The evaluation results demonstrate that the distilled smaller dense models perform exceptionally well on benchmarks. We open-source distilled 1.5B, 7B, 8B, 14B, 32B, and 70B checkpoints based on Qwen2.5 and Llama3 series to the community.
224
+
225
+ ## 3. Model Downloads
226
+
227
+ ### DeepSeek-R1 Models
228
+
229
+ <div align=""center"">
230
+
231
+ | **Model** | **#Total Params** | **#Activated Params** | **Context Length** | **Download** |
232
+ | :------------: | :------------: | :------------: | :------------: | :------------: |
233
+ | DeepSeek-R1-Zero | 671B | 37B | 128K | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero) |
234
+ | DeepSeek-R1 | 671B | 37B | 128K | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1) |
235
+
236
+ </div>
237
+
238
+ DeepSeek-R1-Zero & DeepSeek-R1 are trained based on DeepSeek-V3-Base.
239
+ For more details regarding the model architecture, please refer to [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repository.
240
+
241
+ ### DeepSeek-R1-Distill Models
242
+
243
+ <div align=""center"">
244
+
245
+ | **Model** | **Base Model** | **Download** |
246
+ | :------------: | :------------: | :------------: |
247
+ | DeepSeek-R1-Distill-Qwen-1.5B | [Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) |
248
+ | DeepSeek-R1-Distill-Qwen-7B | [Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-7B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) |
249
+ | DeepSeek-R1-Distill-Llama-8B | [Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) |
250
+ | DeepSeek-R1-Distill-Qwen-14B | [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) |
251
+ |DeepSeek-R1-Distill-Qwen-32B | [Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) |
252
+ | DeepSeek-R1-Distill-Llama-70B | [Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) | [🤗 HuggingFace](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) |
253
+
254
+ </div>
255
+
256
+ DeepSeek-R1-Distill models are fine-tuned based on open-source models, using samples generated by DeepSeek-R1.
257
+ We slightly change their configs and tokenizers. Please use our setting to run these models.
258
+
259
+ ## 4. Evaluation Results
260
+
261
+ ### DeepSeek-R1-Evaluation
262
+ For all our models, the maximum generation length is set to 32,768 tokens. For benchmarks requiring sampling, we use a temperature of $0.6$, a top-p value of $0.95$, and generate 64 responses per query to estimate pass@1.
263
+ <div align=""center"">
264
+
265
+
266
+ | Category | Benchmark (Metric) | Claude-3.5-Sonnet-1022 | GPT-4o 0513 | DeepSeek V3 | OpenAI o1-mini | OpenAI o1-1217 | DeepSeek R1 |
267
+ |----------|-------------------|----------------------|------------|--------------|----------------|------------|--------------|
268
+ | | Architecture | - | - | MoE | - | - | MoE |
269
+ | | # Activated Params | - | - | 37B | - | - | 37B |
270
+ | | # Total Params | - | - | 671B | - | - | 671B |
271
+ | English | MMLU (Pass@1) | 88.3 | 87.2 | 88.5 | 85.2 | **91.8** | 90.8 |
272
+ | | MMLU-Redux (EM) | 88.9 | 88.0 | 89.1 | 86.7 | - | **92.9** |
273
+ | | MMLU-Pro (EM) | 78.0 | 72.6 | 75.9 | 80.3 | - | **84.0** |
274
+ | | DROP (3-shot F1) | 88.3 | 83.7 | 91.6 | 83.9 | 90.2 | **92.2** |
275
+ | | IF-Eval (Prompt Strict) | **86.5** | 84.3 | 86.1 | 84.8 | - | 83.3 |
276
+ | | GPQA-Diamond (Pass@1) | 65.0 | 49.9 | 59.1 | 60.0 | **75.7** | 71.5 |
277
+ | | SimpleQA (Correct) | 28.4 | 38.2 | 24.9 | 7.0 | **47.0** | 30.1 |
278
+ | | FRAMES (Acc.) | 72.5 | 80.5 | 73.3 | 76.9 | - | **82.5** |
279
+ | | AlpacaEval2.0 (LC-winrate) | 52.0 | 51.1 | 70.0 | 57.8 | - | **87.6** |
280
+ | | ArenaHard (GPT-4-1106) | 85.2 | 80.4 | 85.5 | 92.0 | - | **92.3** |
281
+ | Code | LiveCodeBench (Pass@1-COT) | 33.8 | 34.2 | - | 53.8 | 63.4 | **65.9** |
282
+ | | Codeforces (Percentile) | 20.3 | 23.6 | 58.7 | 93.4 | **96.6** | 96.3 |
283
+ | | Codeforces (Rating) | 717 | 759 | 1134 | 1820 | **2061** | 2029 |
284
+ | | SWE Verified (Resolved) | **50.8** | 38.8 | 42.0 | 41.6 | 48.9 | 49.2 |
285
+ | | Aider-Polyglot (Acc.) | 45.3 | 16.0 | 49.6 | 32.9 | **61.7** | 53.3 |
286
+ | Math | AIME 2024 (Pass@1) | 16.0 | 9.3 | 39.2 | 63.6 | 79.2 | **79.8** |
287
+ | | MATH-500 (Pass@1) | 78.3 | 74.6 | 90.2 | 90.0 | 96.4 | **97.3** |
288
+ | | CNMO 2024 (Pass@1) | 13.1 | 10.8 | 43.2 | 67.6 | - | **78.8** |
289
+ | Chinese | CLUEWSC (EM) | 85.4 | 87.9 | 90.9 | 89.9 | - | **92.8** |
290
+ | | C-Eval (EM) | 76.7 | 76.0 | 86.5 | 68.9 | - | **91.8** |
291
+ | | C-SimpleQA (Correct) | 55.4 | 58.7 | **68.0** | 40.3 | - | 63.7 |
292
+
293
+ </div>
294
+
295
+
296
+ ### Distilled Model Evaluation
297
+
298
+
299
+ <div align=""center"">
300
+
301
+ | Model | AIME 2024 pass@1 | AIME 2024 cons@64 | MATH-500 pass@1 | GPQA Diamond pass@1 | LiveCodeBench pass@1 | CodeForces rating |
302
+ |------------------------------------------|------------------|-------------------|-----------------|----------------------|----------------------|-------------------|
303
+ | GPT-4o-0513 | 9.3 | 13.4 | 74.6 | 49.9 | 32.9 | 759 |
304
+ | Claude-3.5-Sonnet-1022 | 16.0 | 26.7 | 78.3 | 65.0 | 38.9 | 717 |
305
+ | o1-mini | 63.6 | 80.0 | 90.0 | 60.0 | 53.8 | **1820** |
306
+ | QwQ-32B-Preview | 44.0 | 60.0 | 90.6 | 54.5 | 41.9 | 1316 |
307
+ | DeepSeek-R1-Distill-Qwen-1.5B | 28.9 | 52.7 | 83.9 | 33.8 | 16.9 | 954 |
308
+ | DeepSeek-R1-Distill-Qwen-7B | 55.5 | 83.3 | 92.8 | 49.1 | 37.6 | 1189 |
309
+ | DeepSeek-R1-Distill-Qwen-14B | 69.7 | 80.0 | 93.9 | 59.1 | 53.1 | 1481 |
310
+ | DeepSeek-R1-Distill-Qwen-32B | **72.6** | 83.3 | 94.3 | 62.1 | 57.2 | 1691 |
311
+ | DeepSeek-R1-Distill-Llama-8B | 50.4 | 80.0 | 89.1 | 49.0 | 39.6 | 1205 |
312
+ | DeepSeek-R1-Distill-Llama-70B | 70.0 | **86.7** | **94.5** | **65.2** | **57.5** | 1633 |
313
+
314
+ </div>
315
+
316
+
317
+ ## 5. Chat Website & API Platform
318
+ You can chat with DeepSeek-R1 on DeepSeek's official website: [chat.deepseek.com](https://chat.deepseek.com), and switch on the button ""DeepThink""
319
+
320
+ We also provide OpenAI-Compatible API at DeepSeek Platform: [platform.deepseek.com](https://platform.deepseek.com/)
321
+
322
+ ## 6. How to Run Locally
323
+
324
+ ### DeepSeek-R1 Models
325
+
326
+ Please visit [DeepSeek-V3](https://github.com/deepseek-ai/DeepSeek-V3) repo for more information about running DeepSeek-R1 locally.
327
+
328
+ ### DeepSeek-R1-Distill Models
329
+
330
+ DeepSeek-R1-Distill models can be utilized in the same manner as Qwen or Llama models.
331
+
332
+ For instance, you can easily start a service using [vLLM](https://github.com/vllm-project/vllm):
333
+
334
+ ```shell
335
+ vllm serve deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --tensor-parallel-size 2 --max-model-len 32768 --enforce-eager
336
+ ```
337
+
338
+ You can also easily start a service using [SGLang](https://github.com/sgl-project/sglang)
339
+
340
+ ```bash
341
+ python3 -m sglang.launch_server --model deepseek-ai/DeepSeek-R1-Distill-Qwen-32B --trust-remote-code --tp 2
342
+ ```
343
+
344
+ ### Usage Recommendations
345
+
346
+ **We recommend adhering to the following configurations when utilizing the DeepSeek-R1 series models, including benchmarking, to achieve the expected performance:**
347
+
348
+ 1. Set the temperature within the range of 0.5-0.7 (0.6 is recommended) to prevent endless repetitions or incoherent outputs.
349
+ 2. **Avoid adding a system prompt; all instructions should be contained within the user prompt.**
350
+ 3. For mathematical problems, it is advisable to include a directive in your prompt such as: ""Please reason step by step, and put your final answer within \boxed{}.""
351
+ 4. When evaluating model performance, it is recommended to conduct multiple tests and average the results.
352
+
353
+ ## 7. License
354
+ This code repository and the model weights are licensed under the [MIT License](https://github.com/deepseek-ai/DeepSeek-R1/blob/main/LICENSE).
355
+ DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that:
356
+ - DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, DeepSeek-R1-Distill-Qwen-14B and DeepSeek-R1-Distill-Qwen-32B are derived from [Qwen-2.5 series](https://github.com/QwenLM/Qwen2.5), which are originally licensed under [Apache 2.0 License](https://huggingface.co/Qwen/Qwen2.5-1.5B/blob/main/LICENSE), and now finetuned with 800k samples curated with DeepSeek-R1.
357
+ - DeepSeek-R1-Distill-Llama-8B is derived from Llama3.1-8B-Base and is originally licensed under [llama3.1 license](https://huggingface.co/meta-llama/Llama-3.1-8B/blob/main/LICENSE).
358
+ - DeepSeek-R1-Distill-Llama-70B is derived from Llama3.3-70B-Instruct and is originally licensed under [llama3.3 license](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct/blob/main/LICENSE).
359
+
360
+ ## 8. Citation
361
+ ```
362
+ @misc{deepseekai2025deepseekr1incentivizingreasoningcapability,
363
+ title={DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning},
364
+ author={DeepSeek-AI and Daya Guo and Dejian Yang and Haowei Zhang and Junxiao Song and Ruoyu Zhang and Runxin Xu and Qihao Zhu and Shirong Ma and Peiyi Wang and Xiao Bi and Xiaokang Zhang and Xingkai Yu and Yu Wu and Z. F. Wu and Zhibin Gou and Zhihong Shao and Zhuoshu Li and Ziyi Gao and Aixin Liu and Bing Xue and Bingxuan Wang and Bochao Wu and Bei Feng and Chengda Lu and Chenggang Zhao and Chengqi Deng and Chenyu Zhang and Chong Ruan and Damai Dai and Deli Chen and Dongjie Ji and Erhang Li and Fangyun Lin and Fucong Dai and Fuli Luo and Guangbo Hao and Guanting Chen and Guowei Li and H. Zhang and Han Bao and Hanwei Xu and Haocheng Wang and Honghui Ding and Huajian Xin and Huazuo Gao and Hui Qu and Hui Li and Jianzhong Guo and Jiashi Li and Jiawei Wang and Jingchang Chen and Jingyang Yuan and Junjie Qiu and Junlong Li and J. L. Cai and Jiaqi Ni and Jian Liang and Jin Chen and Kai Dong and Kai Hu and Kaige Gao and Kang Guan and Kexin Huang and Kuai Yu and Lean Wang and Lecong Zhang and Liang Zhao and Litong Wang and Liyue Zhang and Lei Xu and Leyi Xia and Mingchuan Zhang and Minghua Zhang and Minghui Tang and Meng Li and Miaojun Wang and Mingming Li and Ning Tian and Panpan Huang and Peng Zhang and Qiancheng Wang and Qinyu Chen and Qiushi Du and Ruiqi Ge and Ruisong Zhang and Ruizhe Pan and Runji Wang and R. J. Chen and R. L. Jin and Ruyi Chen and Shanghao Lu and Shangyan Zhou and Shanhuang Chen and Shengfeng Ye and Shiyu Wang and Shuiping Yu and Shunfeng Zhou and Shuting Pan and S. S. Li and Shuang Zhou and Shaoqing Wu and Shengfeng Ye and Tao Yun and Tian Pei and Tianyu Sun and T. Wang and Wangding Zeng and Wanjia Zhao and Wen Liu and Wenfeng Liang and Wenjun Gao and Wenqin Yu and Wentao Zhang and W. L. Xiao and Wei An and Xiaodong Liu and Xiaohan Wang and Xiaokang Chen and Xiaotao Nie and Xin Cheng and Xin Liu and Xin Xie and Xingchao Liu and Xinyu Yang and Xinyuan Li and Xuecheng Su and Xuheng Lin and X. Q. Li and Xiangyue Jin and Xiaojin Shen and Xiaosha Chen and Xiaowen Sun and Xiaoxiang Wang and Xinnan Song and Xinyi Zhou and Xianzu Wang and Xinxia Shan and Y. K. Li and Y. Q. Wang and Y. X. Wei and Yang Zhang and Yanhong Xu and Yao Li and Yao Zhao and Yaofeng Sun and Yaohui Wang and Yi Yu and Yichao Zhang and Yifan Shi and Yiliang Xiong and Ying He and Yishi Piao and Yisong Wang and Yixuan Tan and Yiyang Ma and Yiyuan Liu and Yongqiang Guo and Yuan Ou and Yuduan Wang and Yue Gong and Yuheng Zou and Yujia He and Yunfan Xiong and Yuxiang Luo and Yuxiang You and Yuxuan Liu and Yuyang Zhou and Y. X. Zhu and Yanhong Xu and Yanping Huang and Yaohui Li and Yi Zheng and Yuchen Zhu and Yunxian Ma and Ying Tang and Yukun Zha and Yuting Yan and Z. Z. Ren and Zehui Ren and Zhangli Sha and Zhe Fu and Zhean Xu and Zhenda Xie and Zhengyan Zhang and Zhewen Hao and Zhicheng Ma and Zhigang Yan and Zhiyu Wu and Zihui Gu and Zijia Zhu and Zijun Liu and Zilin Li and Ziwei Xie and Ziyang Song and Zizheng Pan and Zhen Huang and Zhipeng Xu and Zhongyu Zhang and Zhen Zhang},
365
+ year={2025},
366
+ eprint={2501.12948},
367
+ archivePrefix={arXiv},
368
+ primaryClass={cs.CL},
369
+ url={https://arxiv.org/abs/2501.12948},
370
+ }
371
+
372
+ ```
373
+
374
+ ## 9. Contact
375
+ If you have any questions, please raise an issue or contact us at [service@deepseek.com](service@deepseek.com).
376
+ ","{""id"": ""unsloth/DeepSeek-R1-GGUF"", ""author"": ""unsloth"", ""sha"": ""3c4941a68d1c81aee5ab741c8ae7d1a3504ad847"", ""last_modified"": ""2025-04-23 19:21:44+00:00"", ""created_at"": ""2025-01-20 13:09:42+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 1996881, ""downloads_all_time"": null, ""likes"": 1041, ""library_name"": ""transformers"", ""gguf"": {""total"": 671026419200, ""architecture"": ""deepseek2"", ""context_length"": 163840, ""chat_template"": ""{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set ns = namespace(is_first=false, is_tool=false, is_output_first=true, system_prompt='', is_first_sp=true) %}{%- for message in messages %}{%- if message['role'] == 'system' %}{%- if ns.is_first_sp %}{% set ns.system_prompt = ns.system_prompt + message['content'] %}{% set ns.is_first_sp = false %}{%- else %}{% set ns.system_prompt = ns.system_prompt + '\\n\\n' + message['content'] %}{%- endif %}{%- endif %}{%- endfor %}{{ bos_token }}{{ ns.system_prompt }}{%- for message in messages %}{%- if message['role'] == 'user' %}{%- set ns.is_tool = false -%}{{'<\uff5cUser\uff5c>' + message['content']}}{%- endif %}{%- if message['role'] == 'assistant' and 'tool_calls' in message %}{%- set ns.is_tool = false -%}{%- for tool in message['tool_calls'] %}{%- if not ns.is_first %}{%- if message['content'] is none %}{{'<\uff5cAssistant\uff5c><\uff5ctool\u2581calls\u2581begin\uff5c><\uff5ctool\u2581call\u2581begin\uff5c>' + tool['type'] + '<\uff5ctool\u2581sep\uff5c>' + tool['function']['name'] + '\\n' + '```json' + '\\n' + tool['function']['arguments'] + '\\n' + '```' + '<\uff5ctool\u2581call\u2581end\uff5c>'}}{%- else %}{{'<\uff5cAssistant\uff5c>' + message['content'] + '<\uff5ctool\u2581calls\u2581begin\uff5c><\uff5ctool\u2581call\u2581begin\uff5c>' + tool['type'] + '<\uff5ctool\u2581sep\uff5c>' + tool['function']['name'] + '\\n' + '```json' + '\\n' + tool['function']['arguments'] + '\\n' + '```' + '<\uff5ctool\u2581call\u2581end\uff5c>'}}{%- endif %}{%- set ns.is_first = true -%}{%- else %}{{'\\n' + '<\uff5ctool\u2581call\u2581begin\uff5c>' + tool['type'] + '<\uff5ctool\u2581sep\uff5c>' + tool['function']['name'] + '\\n' + '```json' + '\\n' + tool['function']['arguments'] + '\\n' + '```' + '<\uff5ctool\u2581call\u2581end\uff5c>'}}{%- endif %}{%- endfor %}{{'<\uff5ctool\u2581calls\u2581end\uff5c><\uff5cend\u2581of\u2581sentence\uff5c>'}}{%- endif %}{%- if message['role'] == 'assistant' and 'tool_calls' not in message %}{%- if ns.is_tool %}{{'<\uff5ctool\u2581outputs\u2581end\uff5c>' + message['content'] + '<\uff5cend\u2581of\u2581sentence\uff5c>'}}{%- set ns.is_tool = false -%}{%- else %}{% set content = message['content'] %}{% if '</think>' in content %}{% set content = content.split('</think>')[-1] %}{% endif %}{{'<\uff5cAssistant\uff5c>' + content + '<\uff5cend\u2581of\u2581sentence\uff5c>'}}{%- endif %}{%- endif %}{%- if message['role'] == 'tool' %}{%- set ns.is_tool = true -%}{%- if ns.is_output_first %}{{'<\uff5ctool\u2581outputs\u2581begin\uff5c><\uff5ctool\u2581output\u2581begin\uff5c>' + message['content'] + '<\uff5ctool\u2581output\u2581end\uff5c>'}}{%- set ns.is_output_first = false %}{%- else %}{{'<\uff5ctool\u2581output\u2581begin\uff5c>' + message['content'] + '<\uff5ctool\u2581output\u2581end\uff5c>'}}{%- endif %}{%- endif %}{%- endfor -%}{% if ns.is_tool %}{{'<\uff5ctool\u2581outputs\u2581end\uff5c>'}}{% endif %}{% if add_generation_prompt and not ns.is_tool %}{{'<\uff5cAssistant\uff5c><think>\\n'}}{% endif %}"", ""bos_token"": ""<\uff5cbegin\u2581of\u2581sentence\uff5c>"", ""eos_token"": ""<\uff5cend\u2581of\u2581sentence\uff5c>""}, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""gguf"", ""deepseek_v3"", ""text-generation"", ""deepseek"", ""unsloth"", ""custom_code"", ""en"", ""arxiv:2501.12948"", ""base_model:deepseek-ai/DeepSeek-R1"", ""base_model:quantized:deepseek-ai/DeepSeek-R1"", ""license:mit"", ""autotrain_compatible"", ""endpoints_compatible"", ""region:us"", ""conversational""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: deepseek-ai/DeepSeek-R1\nlanguage:\n- en\nlibrary_name: transformers\nlicense: mit\ntags:\n- deepseek\n- unsloth\n- transformers"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": null, ""config"": {""architectures"": [""DeepseekV3ForCausalLM""], ""auto_map"": {""AutoConfig"": ""configuration_deepseek.DeepseekV3Config"", ""AutoModel"": ""modeling_deepseek.DeepseekV3Model"", ""AutoModelForCausalLM"": ""modeling_deepseek.DeepseekV3ForCausalLM""}, ""model_type"": ""deepseek_v3""}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00001-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00002-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00003-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00004-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00005-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00006-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00007-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00008-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00009-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00010-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00011-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00012-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00013-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00014-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00015-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00016-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00017-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00018-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00019-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00020-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00021-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00022-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00023-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00024-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00025-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00026-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00027-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00028-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00029-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='BF16/DeepSeek-R1-BF16-00030-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00001-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00002-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00003-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00004-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00005-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00006-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00007-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00008-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00009-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00010-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00011-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00012-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00013-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00014-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00015-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00016-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00017-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00018-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00019-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00020-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00021-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00022-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00023-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00024-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00025-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00026-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00027-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00028-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00029-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-BF16/DeepSeek-R1.BF16-00030-of-00030.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K/DeepSeek-R1-Q2_K-00001-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K/DeepSeek-R1-Q2_K-00002-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K/DeepSeek-R1-Q2_K-00003-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K/DeepSeek-R1-Q2_K-00004-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K/DeepSeek-R1-Q2_K-00005-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K_L/DeepSeek-R1-Q2_K_L-00001-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K_L/DeepSeek-R1-Q2_K_L-00002-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K_L/DeepSeek-R1-Q2_K_L-00003-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K_L/DeepSeek-R1-Q2_K_L-00004-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K_L/DeepSeek-R1-Q2_K_L-00005-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K_XS/DeepSeek-R1-Q2_K_XS-00001-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K_XS/DeepSeek-R1-Q2_K_XS-00002-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K_XS/DeepSeek-R1-Q2_K_XS-00003-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K_XS/DeepSeek-R1-Q2_K_XS-00004-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q2_K_XS/DeepSeek-R1-Q2_K_XS-00005-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q3_K_M/DeepSeek-R1-Q3_K_M-00001-of-00007.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q3_K_M/DeepSeek-R1-Q3_K_M-00002-of-00007.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q3_K_M/DeepSeek-R1-Q3_K_M-00003-of-00007.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q3_K_M/DeepSeek-R1-Q3_K_M-00004-of-00007.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q3_K_M/DeepSeek-R1-Q3_K_M-00005-of-00007.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q3_K_M/DeepSeek-R1-Q3_K_M-00006-of-00007.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q3_K_M/DeepSeek-R1-Q3_K_M-00007-of-00007.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q4_K_M/DeepSeek-R1-Q4_K_M-00001-of-00009.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q4_K_M/DeepSeek-R1-Q4_K_M-00002-of-00009.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q4_K_M/DeepSeek-R1-Q4_K_M-00003-of-00009.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q4_K_M/DeepSeek-R1-Q4_K_M-00004-of-00009.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q4_K_M/DeepSeek-R1-Q4_K_M-00005-of-00009.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q4_K_M/DeepSeek-R1-Q4_K_M-00006-of-00009.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q4_K_M/DeepSeek-R1-Q4_K_M-00007-of-00009.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q4_K_M/DeepSeek-R1-Q4_K_M-00008-of-00009.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q4_K_M/DeepSeek-R1-Q4_K_M-00009-of-00009.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q5_K_M/DeepSeek-R1-Q5_K_M-00001-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q5_K_M/DeepSeek-R1-Q5_K_M-00002-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q5_K_M/DeepSeek-R1-Q5_K_M-00003-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q5_K_M/DeepSeek-R1-Q5_K_M-00004-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q5_K_M/DeepSeek-R1-Q5_K_M-00005-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q5_K_M/DeepSeek-R1-Q5_K_M-00006-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q5_K_M/DeepSeek-R1-Q5_K_M-00007-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q5_K_M/DeepSeek-R1-Q5_K_M-00008-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q5_K_M/DeepSeek-R1-Q5_K_M-00009-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q5_K_M/DeepSeek-R1-Q5_K_M-00010-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q6_K/DeepSeek-R1-Q6_K-00001-of-00012.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q6_K/DeepSeek-R1-Q6_K-00002-of-00012.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q6_K/DeepSeek-R1-Q6_K-00003-of-00012.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q6_K/DeepSeek-R1-Q6_K-00004-of-00012.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q6_K/DeepSeek-R1-Q6_K-00005-of-00012.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q6_K/DeepSeek-R1-Q6_K-00006-of-00012.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q6_K/DeepSeek-R1-Q6_K-00007-of-00012.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q6_K/DeepSeek-R1-Q6_K-00008-of-00012.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q6_K/DeepSeek-R1-Q6_K-00009-of-00012.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q6_K/DeepSeek-R1-Q6_K-00010-of-00012.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q6_K/DeepSeek-R1-Q6_K-00011-of-00012.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q6_K/DeepSeek-R1-Q6_K-00012-of-00012.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00001-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00002-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00003-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00004-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00005-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00006-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00007-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00008-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00009-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00010-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00011-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00012-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00013-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00014-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-00015-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-IQ1_M/DeepSeek-R1-UD-IQ1_M-00001-of-00004.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-IQ1_M/DeepSeek-R1-UD-IQ1_M-00002-of-00004.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-IQ1_M/DeepSeek-R1-UD-IQ1_M-00003-of-00004.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-IQ1_M/DeepSeek-R1-UD-IQ1_M-00004-of-00004.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-IQ1_S/DeepSeek-R1-UD-IQ1_S-00001-of-00003.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-IQ1_S/DeepSeek-R1-UD-IQ1_S-00002-of-00003.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-IQ1_S/DeepSeek-R1-UD-IQ1_S-00003-of-00003.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-IQ2_XXS/DeepSeek-R1-UD-IQ2_XXS-00001-of-00004.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-IQ2_XXS/DeepSeek-R1-UD-IQ2_XXS-00002-of-00004.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-IQ2_XXS/DeepSeek-R1-UD-IQ2_XXS-00003-of-00004.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-IQ2_XXS/DeepSeek-R1-UD-IQ2_XXS-00004-of-00004.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-Q2_K_XL/DeepSeek-R1-UD-Q2_K_XL-00001-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-Q2_K_XL/DeepSeek-R1-UD-Q2_K_XL-00002-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-Q2_K_XL/DeepSeek-R1-UD-Q2_K_XL-00003-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-Q2_K_XL/DeepSeek-R1-UD-Q2_K_XL-00004-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DeepSeek-R1-UD-Q2_K_XL/DeepSeek-R1-UD-Q2_K_XL-00005-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q2_K_L/DeepSeek-R1-Q2_K_L-00001-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q2_K_L/DeepSeek-R1-Q2_K_L-00002-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q2_K_L/DeepSeek-R1-Q2_K_L-00003-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q2_K_L/DeepSeek-R1-Q2_K_L-00004-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q2_K_L/DeepSeek-R1-Q2_K_L-00005-of-00005.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q3_K_M/DeepSeek-R1-Q3_K_M-00001-of-00007.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q3_K_M/DeepSeek-R1-Q3_K_M-00002-of-00007.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q3_K_M/DeepSeek-R1-Q3_K_M-00003-of-00007.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q3_K_M/DeepSeek-R1-Q3_K_M-00004-of-00007.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q3_K_M/DeepSeek-R1-Q3_K_M-00005-of-00007.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q3_K_M/DeepSeek-R1-Q3_K_M-00006-of-00007.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q3_K_M/DeepSeek-R1-Q3_K_M-00007-of-00007.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q5_K_M/DeepSeek-R1-Q5_K_M-00001-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q5_K_M/DeepSeek-R1-Q5_K_M-00002-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q5_K_M/DeepSeek-R1-Q5_K_M-00003-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q5_K_M/DeepSeek-R1-Q5_K_M-00004-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q5_K_M/DeepSeek-R1-Q5_K_M-00005-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q5_K_M/DeepSeek-R1-Q5_K_M-00006-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q5_K_M/DeepSeek-R1-Q5_K_M-00007-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q5_K_M/DeepSeek-R1-Q5_K_M-00008-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q5_K_M/DeepSeek-R1-Q5_K_M-00009-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q5_K_M/DeepSeek-R1-Q5_K_M-00010-of-00010.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00001-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00002-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00003-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00004-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00005-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00006-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00007-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00008-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00009-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00010-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00011-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00012-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00013-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00014-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Q8_0/DeepSeek-R1-Q8_0-00015-of-00015.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='imatrix_unsloth.dat', size=None, blob_id=None, lfs=None)""], ""spaces"": [""PhillHenry/MyLlmPlayground"", ""krishna-k/deepseek-r1""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-04-23 19:21:44+00:00"", ""cardData"": ""base_model: deepseek-ai/DeepSeek-R1\nlanguage:\n- en\nlibrary_name: transformers\nlicense: mit\ntags:\n- deepseek\n- unsloth\n- transformers"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""_id"": ""678e4b16630463a29b67eb5e"", ""modelId"": ""unsloth/DeepSeek-R1-GGUF"", ""usedStorage"": 9778194944831}",0,"https://huggingface.co/sigjnf/unsloth-DeepSeek-R1-GGUF-merged, https://huggingface.co/dhanugowda2342/knowai, https://huggingface.co/ginghalo/deepseek-r1-1_58bit, https://huggingface.co/kingwin97/yiliao, https://huggingface.co/bap25/bap, https://huggingface.co/anandini7/telugu_finetuned_model, https://huggingface.co/VidhyaN/neuralsorority43",7,https://huggingface.co/diyamanna/bhashinillm,1,,0,,0,"PhillHenry/MyLlmPlayground, huggingface/InferenceSupport/discussions/635, krishna-k/deepseek-r1",3
377
+ sigjnf/unsloth-DeepSeek-R1-GGUF-merged,"---
378
+ license: mit
379
+ base_model:
380
+ - unsloth/DeepSeek-R1-GGUF
381
+ ---","{""id"": ""sigjnf/unsloth-DeepSeek-R1-GGUF-merged"", ""author"": ""sigjnf"", ""sha"": ""0b78dda546a4354c694da6f7e44f8ede0adb05ec"", ""last_modified"": ""2025-01-27 20:29:21+00:00"", ""created_at"": ""2025-01-27 20:28:06+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""base_model:unsloth/DeepSeek-R1-GGUF"", ""base_model:finetune:unsloth/DeepSeek-R1-GGUF"", ""license:mit"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- unsloth/DeepSeek-R1-GGUF\nlicense: mit"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-01-27 20:29:21+00:00"", ""cardData"": ""base_model:\n- unsloth/DeepSeek-R1-GGUF\nlicense: mit"", ""transformersInfo"": null, ""_id"": ""6797ec56ef633ea7ad91f217"", ""modelId"": ""sigjnf/unsloth-DeepSeek-R1-GGUF-merged"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=sigjnf/unsloth-DeepSeek-R1-GGUF-merged&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bsigjnf%2Funsloth-DeepSeek-R1-GGUF-merged%5D(%2Fsigjnf%2Funsloth-DeepSeek-R1-GGUF-merged)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
382
+ dhanugowda2342/knowai,"---
383
+ datasets:
384
+ - fdgvjhb/pennydataset
385
+ - cognitivecomputations/dolphin-r1
386
+ language:
387
+ - en
388
+ base_model:
389
+ - unsloth/DeepSeek-R1-GGUF
390
+ ---","{""id"": ""dhanugowda2342/knowai"", ""author"": ""dhanugowda2342"", ""sha"": ""683e295cef0a80b49ddd569c5450d2cd83c1d053"", ""last_modified"": ""2025-02-02 10:47:39+00:00"", ""created_at"": ""2025-02-02 10:46:22+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""en"", ""dataset:fdgvjhb/pennydataset"", ""dataset:cognitivecomputations/dolphin-r1"", ""base_model:unsloth/DeepSeek-R1-GGUF"", ""base_model:finetune:unsloth/DeepSeek-R1-GGUF"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- unsloth/DeepSeek-R1-GGUF\ndatasets:\n- fdgvjhb/pennydataset\n- cognitivecomputations/dolphin-r1\nlanguage:\n- en"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-02 10:47:39+00:00"", ""cardData"": ""base_model:\n- unsloth/DeepSeek-R1-GGUF\ndatasets:\n- fdgvjhb/pennydataset\n- cognitivecomputations/dolphin-r1\nlanguage:\n- en"", ""transformersInfo"": null, ""_id"": ""679f4cfe661323875caa3dc1"", ""modelId"": ""dhanugowda2342/knowai"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=dhanugowda2342/knowai&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bdhanugowda2342%2Fknowai%5D(%2Fdhanugowda2342%2Fknowai)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
391
+ ginghalo/deepseek-r1-1_58bit,"---
392
+ license: mit
393
+ base_model:
394
+ - unsloth/DeepSeek-R1-GGUF
395
+ ---
396
+ ","{""id"": ""ginghalo/deepseek-r1-1_58bit"", ""author"": ""ginghalo"", ""sha"": ""09ba763656ce63ba83aa7e40248f7476004c974b"", ""last_modified"": ""2025-02-10 03:02:23+00:00"", ""created_at"": ""2025-02-10 03:00:40+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""base_model:unsloth/DeepSeek-R1-GGUF"", ""base_model:finetune:unsloth/DeepSeek-R1-GGUF"", ""license:mit"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- unsloth/DeepSeek-R1-GGUF\nlicense: mit"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-10 03:02:23+00:00"", ""cardData"": ""base_model:\n- unsloth/DeepSeek-R1-GGUF\nlicense: mit"", ""transformersInfo"": null, ""_id"": ""67a96bd830e8962a38889de5"", ""modelId"": ""ginghalo/deepseek-r1-1_58bit"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=ginghalo/deepseek-r1-1_58bit&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bginghalo%2Fdeepseek-r1-1_58bit%5D(%2Fginghalo%2Fdeepseek-r1-1_58bit)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
397
+ kingwin97/yiliao,"---
398
+ license: afl-3.0
399
+ base_model:
400
+ - unsloth/DeepSeek-R1-GGUF
401
+ pipeline_tag: text-generation
402
+ tags:
403
+ - medical
404
+ language:
405
+ - aa
406
+ library_name: fasttext
407
+ ---","{""id"": ""kingwin97/yiliao"", ""author"": ""kingwin97"", ""sha"": ""cda16f88a3704a82d7f8b95a994ae6a7eea2009b"", ""last_modified"": ""2025-02-16 15:19:19+00:00"", ""created_at"": ""2025-02-16 10:55:18+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""fasttext"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""fasttext"", ""medical"", ""text-generation"", ""aa"", ""base_model:unsloth/DeepSeek-R1-GGUF"", ""base_model:finetune:unsloth/DeepSeek-R1-GGUF"", ""license:afl-3.0"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- unsloth/DeepSeek-R1-GGUF\nlanguage:\n- aa\nlibrary_name: fasttext\nlicense: afl-3.0\npipeline_tag: text-generation\ntags:\n- medical"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='Llama3-FP16.ggu', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-16 15:19:19+00:00"", ""cardData"": ""base_model:\n- unsloth/DeepSeek-R1-GGUF\nlanguage:\n- aa\nlibrary_name: fasttext\nlicense: afl-3.0\npipeline_tag: text-generation\ntags:\n- medical"", ""transformersInfo"": null, ""_id"": ""67b1c416d2ee8e627da8b945"", ""modelId"": ""kingwin97/yiliao"", ""usedStorage"": 16063688000}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=kingwin97/yiliao&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bkingwin97%2Fyiliao%5D(%2Fkingwin97%2Fyiliao)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
408
+ bap25/bap,"---
409
+ license: other
410
+ license_name: bap
411
+ license_link: LICENSE
412
+ datasets:
413
+ - FreedomIntelligence/medical-o1-reasoning-SFT
414
+ metrics:
415
+ - bertscore
416
+ base_model:
417
+ - unsloth/DeepSeek-R1-GGUF
418
+ new_version: unsloth/DeepSeek-R1-GGUF
419
+ pipeline_tag: text-to-audio
420
+ library_name: diffusers
421
+ tags:
422
+ - music
423
+ ---","{""id"": ""bap25/bap"", ""author"": ""bap25"", ""sha"": ""154901d8e5800507579d8d07c7ba8665a264ff0e"", ""last_modified"": ""2025-02-18 15:14:59+00:00"", ""created_at"": ""2025-02-18 15:13:53+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""diffusers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""diffusers"", ""music"", ""text-to-audio"", ""dataset:FreedomIntelligence/medical-o1-reasoning-SFT"", ""base_model:unsloth/DeepSeek-R1-GGUF"", ""base_model:finetune:unsloth/DeepSeek-R1-GGUF"", ""license:other"", ""region:us""], ""pipeline_tag"": ""text-to-audio"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- unsloth/DeepSeek-R1-GGUF\ndatasets:\n- FreedomIntelligence/medical-o1-reasoning-SFT\nlibrary_name: diffusers\nlicense: other\nlicense_name: bap\nlicense_link: LICENSE\nmetrics:\n- bertscore\npipeline_tag: text-to-audio\ntags:\n- music\nnew_version: unsloth/DeepSeek-R1-GGUF"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='LICENSE', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-18 15:14:59+00:00"", ""cardData"": ""base_model:\n- unsloth/DeepSeek-R1-GGUF\ndatasets:\n- FreedomIntelligence/medical-o1-reasoning-SFT\nlibrary_name: diffusers\nlicense: other\nlicense_name: bap\nlicense_link: LICENSE\nmetrics:\n- bertscore\npipeline_tag: text-to-audio\ntags:\n- music\nnew_version: unsloth/DeepSeek-R1-GGUF"", ""transformersInfo"": null, ""_id"": ""67b4a3b16919d701a15dde88"", ""modelId"": ""bap25/bap"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=bap25/bap&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bbap25%2Fbap%5D(%2Fbap25%2Fbap)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
424
+ anandini7/telugu_finetuned_model,"---
425
+ license: mit
426
+ language:
427
+ - te
428
+ base_model:
429
+ - unsloth/DeepSeek-R1-GGUF
430
+ new_version: deepseek-ai/DeepSeek-V3
431
+ pipeline_tag: translation
432
+ library_name: diffusers
433
+ tags:
434
+ - code
435
+ ---","{""id"": ""anandini7/telugu_finetuned_model"", ""author"": ""anandini7"", ""sha"": ""d8b31afdffe449282f7a549557abebb3ebad2a48"", ""last_modified"": ""2025-02-19 05:54:55+00:00"", ""created_at"": ""2025-02-18 20:47:51+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""diffusers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""diffusers"", ""code"", ""translation"", ""te"", ""base_model:unsloth/DeepSeek-R1-GGUF"", ""base_model:finetune:unsloth/DeepSeek-R1-GGUF"", ""license:mit"", ""region:us""], ""pipeline_tag"": ""translation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- unsloth/DeepSeek-R1-GGUF\nlanguage:\n- te\nlibrary_name: diffusers\nlicense: mit\npipeline_tag: translation\ntags:\n- code\nnew_version: deepseek-ai/DeepSeek-V3"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-19 05:54:55+00:00"", ""cardData"": ""base_model:\n- unsloth/DeepSeek-R1-GGUF\nlanguage:\n- te\nlibrary_name: diffusers\nlicense: mit\npipeline_tag: translation\ntags:\n- code\nnew_version: deepseek-ai/DeepSeek-V3"", ""transformersInfo"": null, ""_id"": ""67b4f1f7871574cd38eebf40"", ""modelId"": ""anandini7/telugu_finetuned_model"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=anandini7/telugu_finetuned_model&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Banandini7%2Ftelugu_finetuned_model%5D(%2Fanandini7%2Ftelugu_finetuned_model)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
436
+ VidhyaN/neuralsorority43,"---
437
+ license: apache-2.0
438
+ base_model:
439
+ - unsloth/DeepSeek-R1-GGUF
440
+ tags:
441
+ - code
442
+ ---","{""id"": ""VidhyaN/neuralsorority43"", ""author"": ""VidhyaN"", ""sha"": ""9976cc8ada11b9acdaa9b2d8f298bf81fbefb063"", ""last_modified"": ""2025-02-20 17:16:23+00:00"", ""created_at"": ""2025-02-20 17:10:42+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""code"", ""base_model:unsloth/DeepSeek-R1-GGUF"", ""base_model:finetune:unsloth/DeepSeek-R1-GGUF"", ""license:apache-2.0"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- unsloth/DeepSeek-R1-GGUF\nlicense: apache-2.0\ntags:\n- code"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='translate_project.ipynb', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-20 17:16:23+00:00"", ""cardData"": ""base_model:\n- unsloth/DeepSeek-R1-GGUF\nlicense: apache-2.0\ntags:\n- code"", ""transformersInfo"": null, ""_id"": ""67b762125331405aee2904b1"", ""modelId"": ""VidhyaN/neuralsorority43"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=VidhyaN/neuralsorority43&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BVidhyaN%2Fneuralsorority43%5D(%2FVidhyaN%2Fneuralsorority43)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
Flux-Super-Realism-LoRA_finetunes_20250426_215237.csv_finetunes_20250426_215237.csv ADDED
@@ -0,0 +1,151 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ strangerzonehf/Flux-Super-Realism-LoRA,"---
3
+ tags:
4
+ - text-to-image
5
+ - lora
6
+ - diffusers
7
+ - template:diffusion-lora
8
+ - Super-Realism
9
+ - Flux.1-Dev
10
+ - Dynamic-Realism
11
+ - Realistic
12
+ - Photorealism
13
+ - Hi-Res
14
+ - UltraRealism
15
+ - Diffusion
16
+ - Face
17
+ - safetensors
18
+ - Realism-Engine
19
+ - RAW
20
+ - 4K
21
+ widget:
22
+ - text: >-
23
+ Super Realism, Woman in a red jacket, snowy, in the style of hyper-realistic
24
+ portraiture, caninecore, mountainous vistas, timeless beauty, palewave,
25
+ iconic, distinctive noses --ar 72:101 --stylize 750 --v 6
26
+ output:
27
+ url: images/3.png
28
+ - text: >-
29
+ Super Realism, Headshot of handsome young man, wearing dark gray sweater
30
+ with buttons and big shawl collar, brown hair and short beard, serious look
31
+ on his face, black background, soft studio lighting, portrait photography
32
+ --ar 85:128 --v 6.0 --style rawHeadshot of handsome young man, wearing dark
33
+ gray sweater with buttons and big shawl collar, brown hair and short beard,
34
+ serious look on his face, black background, soft studio lighting, portrait
35
+ photography --ar 85:128 --v 6.0 --style rawHeadshot of handsome young man,
36
+ wearing dark gray sweater with buttons and big shawl collar, brown hair and
37
+ short beard, serious look on his face, black background, soft studio
38
+ lighting, portrait photography --ar 85:128 --v 6.0 --style raw
39
+ output:
40
+ url: images/2.png
41
+ - text: >-
42
+ Super Realism, High-resolution photograph, woman, UHD, photorealistic, shot
43
+ on a Sony A7III --chaos 20 --ar 1:2 --style raw --stylize 250
44
+ output:
45
+ url: images/1.png
46
+ base_model: black-forest-labs/FLUX.1-dev
47
+ instance_prompt: Super Realism
48
+ license: mit
49
+ ---
50
+ ![strangerzonehf/Flux-Super-Realism-LoRA](images/sz.png)
51
+
52
+ <Gallery />
53
+
54
+ ## Model description for super realism engine
55
+
56
+ Image Processing Parameters
57
+
58
+ | Parameter | Value | Parameter | Value |
59
+ |---------------------------|--------|---------------------------|--------|
60
+ | LR Scheduler | constant | Noise Offset | 0.03 |
61
+ | Optimizer | AdamW | Multires Noise Discount | 0.1 |
62
+ | Network Dim | 64 | Multires Noise Iterations | 10 |
63
+ | Network Alpha | 32 | Repeat & Steps | 30 & 4380|
64
+ | Epoch | 20 | Save Every N Epochs | 1 |
65
+
66
+ ## Comparison between the base model and related models.
67
+
68
+ Comparison between the base model FLUX.1-dev and its adapter, a LoRA model tuned for super-realistic realism.
69
+ [ 28 steps ]
70
+
71
+ ![strangerzonehf/Flux-Super-Realism-LoRA](images/sz2.png)
72
+
73
+ However, it performs better in various aspects compared to its previous models, including face realism, ultra-realism, and others.
74
+ previous versions [ 28 steps ]
75
+
76
+ ![strangerzonehf/Flux-Super-Realism-LoRA](images/sz3.png)
77
+
78
+ ## Previous Model Links
79
+
80
+ | Model Name | Description | Link |
81
+ |------------------------------------------|------------------------------|------------------------------------------------------------------------------------------|
82
+ | **Canopus-LoRA-Flux-FaceRealism** | LoRA model for Face Realism | [Canopus-LoRA-Flux-FaceRealism](https://huggingface.co/prithivMLmods/Canopus-LoRA-Flux-FaceRealism) |
83
+ | **Canopus-LoRA-Flux-UltraRealism-2.0** | LoRA model for Ultra Realism | [Canopus-LoRA-Flux-UltraRealism-2.0](https://huggingface.co/prithivMLmods/Canopus-LoRA-Flux-UltraRealism-2.0) |
84
+ | **Flux.1-Dev-LoRA-HDR-Realism [Experimental Version]** | LoRA model for HDR Realism | [Flux.1-Dev-LoRA-HDR-Realism](https://huggingface.co/prithivMLmods/Flux.1-Dev-LoRA-HDR-Realism) |
85
+ | **Flux-Realism-FineDetailed** | Fine-detailed realism-focused model | [Flux-Realism-FineDetailed](https://huggingface.co/prithivMLmods/Flux-Realism-FineDetailed) |
86
+
87
+ ## Hosted/Demo Links
88
+
89
+ | Demo Name | Description | Link |
90
+ |----------------------------|----------------------------|--------------------------------------------------------------------------------------|
91
+ | **FLUX-LoRA-DLC** | Demo for FLUX LoRA DLC | [FLUX-LoRA-DLC](https://huggingface.co/spaces/prithivMLmods/FLUX-LoRA-DLC) |
92
+ | **FLUX-REALISM** | Demo for FLUX Realism | [FLUX-REALISM](https://huggingface.co/spaces/prithivMLmods/FLUX-REALISM) |
93
+
94
+ ## Model Training Basic Details
95
+
96
+ | Feature | Description |
97
+ |--------------------------------|-------------------------------------|
98
+ | **Labeling** | florence2-en (natural language & English) |
99
+ | **Total Images Used for Training** | 55 [Hi-Res] |
100
+ | **Best Dimensions** | - 1024 x 1024 (Default) |
101
+ | | - 768 x 1024 |
102
+
103
+ ## Flux-Super-Realism-LoRA Model GitHub
104
+
105
+ | Repository Link | Description |
106
+ |---------------------------------------------------------------|----------------------------------------------|
107
+ | [Flux-Super-Realism-LoRA](https://github.com/Stranger-Zone/Flux-Super-Realism-LoRA/tree/main) | Flux Super Realism LoRA model repository for high-quality realism generation |
108
+
109
+ ## API Usage / Quick Usage
110
+ ```python
111
+ from gradio_client import Client
112
+
113
+ client = Client(""prithivMLmods/FLUX-REALISM"")
114
+ result = client.predict(
115
+ prompt=""A tiny astronaut hatching from an egg on the moon, 4k, planet theme"",
116
+ seed=0,
117
+ width=1024,
118
+ height=1024,
119
+ guidance_scale=6,
120
+ randomize_seed=True,
121
+ api_name=""/run""
122
+ #takes minimum of 30 seconds
123
+ )
124
+ print(result)
125
+ ```
126
+ ## Setting Up Flux Space
127
+ ```python
128
+ import torch
129
+ from pipelines import DiffusionPipeline
130
+
131
+ base_model = ""black-forest-labs/FLUX.1-dev""
132
+ pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
133
+
134
+ lora_repo = ""strangerzonehf/Flux-Super-Realism-LoRA""
135
+ trigger_word = ""Super Realism"" #triggerword
136
+ pipe.load_lora_weights(lora_repo)
137
+
138
+ device = torch.device(""cuda"")
139
+ pipe.to(device)
140
+ ```
141
+ ## Trigger words
142
+
143
+ > [!WARNING]
144
+ > **Trigger words:** You should use `Super Realism` to trigger the image generation.
145
+
146
+ - The trigger word is not mandatory; ensure that words like ""realistic"" and ""realism"" appear in the image description. The ""super realism"" trigger word should prompt an exact match to the reference image in the showcase.
147
+ ## Download model
148
+
149
+ Weights for this model are available in Safetensors format.
150
+
151
+ [Download](/strangerzonehf/Flux-Super-Realism-LoRA/tree/main) them in the Files & versions tab.","{""id"": ""strangerzonehf/Flux-Super-Realism-LoRA"", ""author"": ""strangerzonehf"", ""sha"": ""321693353a5c05fe013ad431578d97e866159515"", ""last_modified"": ""2024-11-27 20:50:39+00:00"", ""created_at"": ""2024-11-12 18:40:16+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 12799, ""downloads_all_time"": null, ""likes"": 202, ""library_name"": ""diffusers"", ""gguf"": null, ""inference"": ""warm"", ""inference_provider_mapping"": null, ""tags"": [""diffusers"", ""text-to-image"", ""lora"", ""template:diffusion-lora"", ""Super-Realism"", ""Flux.1-Dev"", ""Dynamic-Realism"", ""Realistic"", ""Photorealism"", ""Hi-Res"", ""UltraRealism"", ""Diffusion"", ""Face"", ""safetensors"", ""Realism-Engine"", ""RAW"", ""4K"", ""base_model:black-forest-labs/FLUX.1-dev"", ""base_model:adapter:black-forest-labs/FLUX.1-dev"", ""license:mit"", ""region:us""], ""pipeline_tag"": ""text-to-image"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: black-forest-labs/FLUX.1-dev\nlicense: mit\ntags:\n- text-to-image\n- lora\n- diffusers\n- template:diffusion-lora\n- Super-Realism\n- Flux.1-Dev\n- Dynamic-Realism\n- Realistic\n- Photorealism\n- Hi-Res\n- UltraRealism\n- Diffusion\n- Face\n- safetensors\n- Realism-Engine\n- RAW\n- 4K\nwidget:\n- text: Super Realism, Woman in a red jacket, snowy, in the style of hyper-realistic\n portraiture, caninecore, mountainous vistas, timeless beauty, palewave, iconic,\n distinctive noses --ar 72:101 --stylize 750 --v 6\n output:\n url: https://huggingface.co/strangerzonehf/Flux-Super-Realism-LoRA/resolve/main/images/3.png\n- text: Super Realism, Headshot of handsome young man, wearing dark gray sweater with\n buttons and big shawl collar, brown hair and short beard, serious look on his\n face, black background, soft studio lighting, portrait photography --ar 85:128\n --v 6.0 --style rawHeadshot of handsome young man, wearing dark gray sweater with\n buttons and big shawl collar, brown hair and short beard, serious look on his\n face, black background, soft studio lighting, portrait photography --ar 85:128\n --v 6.0 --style rawHeadshot of handsome young man, wearing dark gray sweater with\n buttons and big shawl collar, brown hair and short beard, serious look on his\n face, black background, soft studio lighting, portrait photography --ar 85:128\n --v 6.0 --style raw\n output:\n url: https://huggingface.co/strangerzonehf/Flux-Super-Realism-LoRA/resolve/main/images/2.png\n- text: Super Realism, High-resolution photograph, woman, UHD, photorealistic, shot\n on a Sony A7III --chaos 20 --ar 1:2 --style raw --stylize 250\n output:\n url: https://huggingface.co/strangerzonehf/Flux-Super-Realism-LoRA/resolve/main/images/1.png\ninstance_prompt: Super Realism"", ""widget_data"": [{""text"": ""Super Realism, Woman in a red jacket, snowy, in the style of hyper-realistic portraiture, caninecore, mountainous vistas, timeless beauty, palewave, iconic, distinctive noses --ar 72:101 --stylize 750 --v 6"", ""output"": {""url"": ""https://huggingface.co/strangerzonehf/Flux-Super-Realism-LoRA/resolve/main/images/3.png""}}, {""text"": ""Super Realism, Headshot of handsome young man, wearing dark gray sweater with buttons and big shawl collar, brown hair and short beard, serious look on his face, black background, soft studio lighting, portrait photography --ar 85:128 --v 6.0 --style rawHeadshot of handsome young man, wearing dark gray sweater with buttons and big shawl collar, brown hair and short beard, serious look on his face, black background, soft studio lighting, portrait photography --ar 85:128 --v 6.0 --style rawHeadshot of handsome young man, wearing dark gray sweater with buttons and big shawl collar, brown hair and short beard, serious look on his face, black background, soft studio lighting, portrait photography --ar 85:128 --v 6.0 --style raw"", ""output"": {""url"": ""https://huggingface.co/strangerzonehf/Flux-Super-Realism-LoRA/resolve/main/images/2.png""}}, {""text"": ""Super Realism, High-resolution photograph, woman, UHD, photorealistic, shot on a Sony A7III --chaos 20 --ar 1:2 --style raw --stylize 250"", ""output"": {""url"": ""https://huggingface.co/strangerzonehf/Flux-Super-Realism-LoRA/resolve/main/images/1.png""}}], ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='images/1.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='images/2.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='images/3.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='images/sz.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='images/sz2.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='images/sz3.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='super-realism.safetensors', size=None, blob_id=None, lfs=None)""], ""spaces"": [""prithivMLmods/FLUX-LoRA-DLC"", ""Nymbo/Serverless-ImgGen-Hub"", ""prithivMLmods/FLUX-REALISM"", ""NeurixYUFI/imggen"", ""fantaxy/playground25"", ""Novaciano/Flux_Lustly_AI_Uncensored_NSFW_V1"", ""vyloup/FLUX-LoRA-DLC"", ""guardiancc/FLUX-LoRA-DLC-fixed"", ""ruslanmv/Flux-LoRA-Generation-Advanced"", ""Smiley0707/FLUX-LoRA-DLC"", ""soiz1/FLUX-LoRA-DLC"", ""mantrakp/AllFlux"", ""Dagfinn1962/FLUX-REALISM"", ""revittapanda/Revitta-Flux-Super-Realism-LoRA"", ""cutifly/strangerzonehf-Flux-Super-Realism-LoRA"", ""NativeAngels/Serverless-ImgGen-Hub"", ""NASSIM55/strangerzonehf-Flux-Super-Realism-LoRA"", ""sajjaddg/strangerzonehf-Flux-Super-Realism-LoRA"", ""theunseenones94/Flux_Lustly_AI_Uncensored_NSFW_V1"", ""yufiru/ImageGeneratotModels"", ""guardiancc/arcane"", ""openfree/tesfdsfsf"", ""Nymbo/Model-Status-Checker"", ""crazyhite001/imggen"", ""DJStomp/FLUX-LoRA-DLC"", ""Fabyler/strangerzonehf-Flux-Super-Realism-LoRA"", ""joeysaada/strangerzonehf-Flux-Super-Realism-LoRAdz"", ""Parmist/strangerzonehf-Flux-Super-Realism-LoRA"", ""Kidbea/Kidbea_Image_Generation"", ""Kidbea/multimodels_image_generation"", ""shomare01/Flux-Super-Realism-LoRA"", ""K00B404/FLUX-Wallpaper-HD-Maker_p"", ""Nymbo/serverless-imggen-test"", ""cngsm/FLUX-LoRA-DLC"", ""incude/strangerzonehf-Flux-Super-Realism-LoRA"", ""Xach35/FLUX-LoRA-DLC"", ""Delta-4/Delta4-Super-Realism-LoRA"", ""Nightwing25/strangerzonehf-Flux-Super-Realism-LoRA"", ""soharab/Super-Realism-LoRA"", ""wingsss/strangerzonehf-Flux-Super-Realism-LoRA"", ""ryan171088/FLUX-LoRA-DLC"", ""K00B404/TheUnclonable"", ""Akshit2606/strangerzonehf-Flux-Super-Realism-LoRA"", ""TimHortonsRAW/FLUX-REALISM"", ""nkargar1356/strangerzonehf-Flux-Super-Realism-LoRA"", ""martynka/for-dev"", ""fourmyfriends/FLUX-LoRA-DLC"", ""EliteGamerCJ/strangerzonehf-Flux-Super-Realism-LoRA"", ""njavidfar/o"", ""indianace/strangerzonehf-Flux-Super-Realism-LoRA"", ""bluenevus/picture-perfect"", ""Arashpey/FLUX-LoRA-DLC"", ""codermert/hmmm"", ""drago33drago/FLUX-LoRA-DLC"", ""rafaelkamp/black-forest-labs-FLUX.1-dev""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-11-27 20:50:39+00:00"", ""cardData"": ""base_model: black-forest-labs/FLUX.1-dev\nlicense: mit\ntags:\n- text-to-image\n- lora\n- diffusers\n- template:diffusion-lora\n- Super-Realism\n- Flux.1-Dev\n- Dynamic-Realism\n- Realistic\n- Photorealism\n- Hi-Res\n- UltraRealism\n- Diffusion\n- Face\n- safetensors\n- Realism-Engine\n- RAW\n- 4K\nwidget:\n- text: Super Realism, Woman in a red jacket, snowy, in the style of hyper-realistic\n portraiture, caninecore, mountainous vistas, timeless beauty, palewave, iconic,\n distinctive noses --ar 72:101 --stylize 750 --v 6\n output:\n url: https://huggingface.co/strangerzonehf/Flux-Super-Realism-LoRA/resolve/main/images/3.png\n- text: Super Realism, Headshot of handsome young man, wearing dark gray sweater with\n buttons and big shawl collar, brown hair and short beard, serious look on his\n face, black background, soft studio lighting, portrait photography --ar 85:128\n --v 6.0 --style rawHeadshot of handsome young man, wearing dark gray sweater with\n buttons and big shawl collar, brown hair and short beard, serious look on his\n face, black background, soft studio lighting, portrait photography --ar 85:128\n --v 6.0 --style rawHeadshot of handsome young man, wearing dark gray sweater with\n buttons and big shawl collar, brown hair and short beard, serious look on his\n face, black background, soft studio lighting, portrait photography --ar 85:128\n --v 6.0 --style raw\n output:\n url: https://huggingface.co/strangerzonehf/Flux-Super-Realism-LoRA/resolve/main/images/2.png\n- text: Super Realism, High-resolution photograph, woman, UHD, photorealistic, shot\n on a Sony A7III --chaos 20 --ar 1:2 --style raw --stylize 250\n output:\n url: https://huggingface.co/strangerzonehf/Flux-Super-Realism-LoRA/resolve/main/images/1.png\ninstance_prompt: Super Realism"", ""transformersInfo"": null, ""_id"": ""6733a110b6daf0fbdb660971"", ""modelId"": ""strangerzonehf/Flux-Super-Realism-LoRA"", ""usedStorage"": 644001442}",0,,0,,0,,0,,0,"NeurixYUFI/imggen, Novaciano/Flux_Lustly_AI_Uncensored_NSFW_V1, Nymbo/Serverless-ImgGen-Hub, Smiley0707/FLUX-LoRA-DLC, fantaxy/playground25, guardiancc/FLUX-LoRA-DLC-fixed, mantrakp/AllFlux, prithivMLmods/FLUX-LoRA-DLC, prithivMLmods/FLUX-REALISM, ruslanmv/Flux-LoRA-Generation-Advanced, soiz1/FLUX-LoRA-DLC, vyloup/FLUX-LoRA-DLC",12
GR00T-N1-2B_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ nvidia/GR00T-N1-2B,"---
3
+ datasets:
4
+ - nvidia/PhysicalAI-Robotics-GR00T-X-Embodiment-Sim
5
+ tags:
6
+ - robotics
7
+ ---
8
+
9
+ # GR00T-N1-2B
10
+
11
+
12
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/67b8da81d01134f89899b4a7/8bFQa2ZIGCsOQQ2ho2N_U.png)
13
+
14
+ Github page: https://github.com/NVIDIA/Isaac-GR00T/
15
+
16
+ NVIDIA Isaac GR00T N1 is the world's first open foundation model for generalized humanoid robot reasoning and skills.","{""id"": ""nvidia/GR00T-N1-2B"", ""author"": ""nvidia"", ""sha"": ""32e1fd2507f7739fad443e6b449c8188e0e02fcb"", ""last_modified"": ""2025-03-18 18:35:00+00:00"", ""created_at"": ""2025-03-05 08:40:53+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 3690, ""downloads_all_time"": null, ""likes"": 283, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""safetensors"", ""gr00t_n1"", ""robotics"", ""dataset:nvidia/PhysicalAI-Robotics-GR00T-X-Embodiment-Sim"", ""region:us""], ""pipeline_tag"": ""robotics"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""datasets:\n- nvidia/PhysicalAI-Robotics-GR00T-X-Embodiment-Sim\ntags:\n- robotics"", ""widget_data"": null, ""model_index"": null, ""config"": {""architectures"": [""GR00T_N1""], ""model_type"": ""gr00t_n1""}, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='LICENSE', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='experiment_cfg/metadata.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": {""parameters"": {""BF16"": 2190019826}, ""total"": 2190019826}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-03-18 18:35:00+00:00"", ""cardData"": ""datasets:\n- nvidia/PhysicalAI-Robotics-GR00T-X-Embodiment-Sim\ntags:\n- robotics"", ""transformersInfo"": null, ""_id"": ""67c80e155e535e9b2dca27c2"", ""modelId"": ""nvidia/GR00T-N1-2B"", ""usedStorage"": 13355965552}",0,,0,"https://huggingface.co/ibru/bobo_groot_model_2, https://huggingface.co/ibru/bobo_groot_model_3, https://huggingface.co/ibru/bobo_groot_n1, https://huggingface.co/ibru/bobo_jetson_gr00t_n1_2b, https://huggingface.co/ibru/bob_gr00t_n1_2b_10k, https://huggingface.co/ibru/bob_jetson_gr00t_n1_2b_191e, https://huggingface.co/ibru/bobo_jetson_tennis_gr00t_n1_2b",7,,0,,0,,0
Hunyuan3D-2_finetunes_20250424_223250.csv_finetunes_20250424_223250.csv ADDED
@@ -0,0 +1,459 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ tencent/Hunyuan3D-2,"---
3
+ library_name: hunyuan3d-2
4
+ license: other
5
+ license_name: tencent-hunyuan-community
6
+ license_link: https://huggingface.co/tencent/Hunyuan3D-2/blob/main/LICENSE.txt
7
+ language:
8
+ - en
9
+ - zh
10
+ tags:
11
+ - image-to-3d
12
+ - text-to-3d
13
+ pipeline_tag: image-to-3d
14
+ ---
15
+
16
+ <p align=""center"">
17
+ <img src=""./assets/images/teaser.jpg"">
18
+ </p>
19
+
20
+ <div align=""center"">
21
+ <a href=https://3d.hunyuan.tencent.com target=""_blank""><img src=https://img.shields.io/badge/Hunyuan3D-black.svg?logo=homepage height=22px></a>
22
+ <a href=https://huggingface.co/spaces/tencent/Hunyuan3D-2 target=""_blank""><img src=https://img.shields.io/badge/%F0%9F%A4%97%20Demo-276cb4.svg height=22px></a>
23
+ <a href=https://huggingface.co/tencent/Hunyuan3D-2 target=""_blank""><img src=https://img.shields.io/badge/%F0%9F%A4%97%20Models-d96902.svg height=22px></a>
24
+ <a href=https://3d-models.hunyuan.tencent.com/ target=""_blank""><img src= https://img.shields.io/badge/Page-bb8a2e.svg?logo=github height=22px></a>
25
+ <a href=https://discord.gg/GuaWYwzKbX target=""_blank""><img src= https://img.shields.io/badge/Discord-white.svg?logo=discord height=22px></a>
26
+ <a href=https://github.com/Tencent/Hunyuan3D-2/blob/main/assets/report/Tencent_Hunyuan3D_2_0.pdf target=""_blank""><img src=https://img.shields.io/badge/Report-b5212f.svg?logo=arxiv height=22px></a>
27
+ </div>
28
+
29
+
30
+ [//]: # ( <a href=# target=""_blank""><img src=https://img.shields.io/badge/Report-b5212f.svg?logo=arxiv height=22px></a>)
31
+
32
+ [//]: # ( <a href=# target=""_blank""><img src= https://img.shields.io/badge/Colab-8f2628.svg?logo=googlecolab height=22px></a>)
33
+
34
+ [//]: # ( <a href=""#""><img alt=""PyPI - Downloads"" src=""https://img.shields.io/pypi/v/mulankit?logo=pypi"" height=22px></a>)
35
+
36
+ <br>
37
+ <p align=""center"">
38
+ “ Living out everyone’s imagination on creating and manipulating 3D assets.”
39
+ </p>
40
+
41
+ This repository contains the models of the paper [Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation](https://huggingface.co/papers/2501.12202).
42
+ For code and more details on how to use it, refer to the [Github repository](https://github.com/Tencent/Hunyuan3D-2).
43
+
44
+ ## 🔥 News
45
+
46
+ - Jan 21, 2025: 💬 Release [Hunyuan3D 2.0](https://huggingface.co/spaces/tencent/Hunyuan3D-2). Please give it a try!
47
+
48
+ ## **Abstract**
49
+
50
+ We present Hunyuan3D 2.0, an advanced large-scale 3D synthesis system for generating high-resolution textured 3D assets.
51
+ This system includes two foundation components: a large-scale shape generation model - Hunyuan3D-DiT, and a large-scale
52
+ texture synthesis model - Hunyuan3D-Paint.
53
+ The shape generative model, built on a scalable flow-based diffusion transformer, aims to create geometry that properly
54
+ aligns with a given condition image, laying a solid foundation for downstream applications.
55
+ The texture synthesis model, benefiting from strong geometric and diffusion priors, produces high-resolution and vibrant
56
+ texture maps for either generated or hand-crafted meshes.
57
+ Furthermore, we build Hunyuan3D-Studio - a versatile, user-friendly production platform that simplifies the re-creation
58
+ process of 3D assets. It allows both professional and amateur users to manipulate or even animate their meshes
59
+ efficiently.
60
+ We systematically evaluate our models, showing that Hunyuan3D 2.0 outperforms previous state-of-the-art models,
61
+ including the open-source models and closed-source models in geometry details, condition alignment, texture quality, and
62
+ e.t.c.
63
+
64
+ <p align=""center"">
65
+ <img src=""assets/images/system.jpg"">
66
+ </p>
67
+
68
+ ## ☯️ **Hunyuan3D 2.0**
69
+
70
+ ### Architecture
71
+
72
+ Hunyuan3D 2.0 features a two-stage generation pipeline, starting with the creation of a bare mesh, followed by the
73
+ synthesis of a texture map for that mesh. This strategy is effective for decoupling the difficulties of shape and
74
+ texture generation and also provides flexibility for texturing either generated or handcrafted meshes.
75
+
76
+ <p align=""left"">
77
+ <img src=""assets/images/arch.jpg"">
78
+ </p>
79
+
80
+ ### Performance
81
+
82
+ We have evaluated Hunyuan3D 2.0 with other open-source as well as close-source 3d-generation methods.
83
+ The numerical results indicate that Hunyuan3D 2.0 surpasses all baselines in the quality of generated textured 3D assets
84
+ and the condition following ability.
85
+
86
+ | Model | CMMD(⬇) | FID_CLIP(⬇) | FID(⬇) | CLIP-score(⬆) |
87
+ |-------------------------|-----------|-------------|-------------|---------------|
88
+ | Top Open-source Model1 | 3.591 | 54.639 | 289.287 | 0.787 |
89
+ | Top Close-source Model1 | 3.600 | 55.866 | 305.922 | 0.779 |
90
+ | Top Close-source Model2 | 3.368 | 49.744 | 294.628 | 0.806 |
91
+ | Top Close-source Model3 | 3.218 | 51.574 | 295.691 | 0.799 |
92
+ | Hunyuan3D 2.0 | **3.193** | **49.165** | **282.429** | **0.809** |
93
+
94
+ Generation results of Hunyuan3D 2.0:
95
+ <p align=""left"">
96
+ <img src=""assets/images/e2e-1.gif"" height=300>
97
+ <img src=""assets/images/e2e-2.gif"" height=300>
98
+ </p>
99
+
100
+ ### Pretrained Models
101
+
102
+ | Model | Date | Huggingface |
103
+ |----------------------|------------|--------------------------------------------------------|
104
+ | Hunyuan3D-DiT-v2-0 | 2025-01-21 | [Download](https://huggingface.co/tencent/Hunyuan3D-2) |
105
+ | Hunyuan3D-Paint-v2-0 | 2025-01-21 | [Download](https://huggingface.co/tencent/Hunyuan3D-2) |
106
+ | Hunyuan3D-Delight-v2-0 | 2025-01-21 | [Download](https://huggingface.co/tencent/Hunyuan3D-2/tree/main/hunyuan3d-delight-v2-0) |
107
+
108
+ ## 🤗 Get Started with Hunyuan3D 2.0
109
+
110
+ You may follow the next steps to use Hunyuan3D 2.0 via code or the Gradio App.
111
+
112
+ ### Install Requirements
113
+
114
+ Please install Pytorch via the [official](https://pytorch.org/) site. Then install the other requirements via
115
+
116
+ ```bash
117
+ pip install -r requirements.txt
118
+ # for texture
119
+ cd hy3dgen/texgen/custom_rasterizer
120
+ python3 setup.py install
121
+ cd ../../..
122
+ cd hy3dgen/texgen/differentiable_renderer
123
+ bash compile_mesh_painter.sh OR python3 setup.py install (on Windows)
124
+ ```
125
+
126
+ ### API Usage
127
+
128
+ We designed a diffusers-like API to use our shape generation model - Hunyuan3D-DiT and texture synthesis model -
129
+ Hunyuan3D-Paint.
130
+
131
+ You could assess **Hunyuan3D-DiT** via:
132
+
133
+ ```python
134
+ from hy3dgen.shapegen import Hunyuan3DDiTFlowMatchingPipeline
135
+
136
+ pipeline = Hunyuan3DDiTFlowMatchingPipeline.from_pretrained('tencent/Hunyuan3D-2')
137
+ mesh = pipeline(image='assets/demo.png')[0]
138
+ ```
139
+
140
+ The output mesh is a [trimesh object](https://trimesh.org/trimesh.html), which you could save to glb/obj (or other
141
+ format) file.
142
+
143
+ For **Hunyuan3D-Paint**, do the following:
144
+
145
+ ```python
146
+ from hy3dgen.texgen import Hunyuan3DPaintPipeline
147
+ from hy3dgen.shapegen import Hunyuan3DDiTFlowMatchingPipeline
148
+
149
+ # let's generate a mesh first
150
+ pipeline = Hunyuan3DDiTFlowMatchingPipeline.from_pretrained('tencent/Hunyuan3D-2')
151
+ mesh = pipeline(image='assets/demo.png')[0]
152
+
153
+ pipeline = Hunyuan3DPaintPipeline.from_pretrained('tencent/Hunyuan3D-2')
154
+ mesh = pipeline(mesh, image='assets/demo.png')
155
+ ```
156
+
157
+ Please visit [minimal_demo.py](https://github.com/Tencent/Hunyuan3D-2/blob/main/minimal_demo.py) for more advanced usage, such as **text to 3D** and **texture generation
158
+ for handcrafted mesh**.
159
+
160
+ ### Gradio App
161
+
162
+ You could also host a [Gradio](https://www.gradio.app/) App in your own computer via:
163
+
164
+ ```bash
165
+ pip3 install gradio==3.39.0
166
+ python3 gradio_app.py
167
+ ```
168
+
169
+ Don't forget to visit [Hunyuan3D](https://3d.hunyuan.tencent.com) for quick use, if you don't want to host yourself.
170
+
171
+ ## 📑 Open-Source Plan
172
+
173
+ - [x] Inference Code
174
+ - [x] Model Checkpoints
175
+ - [x] Technical Report
176
+ - [ ] ComfyUI
177
+ - [ ] TensorRT Version
178
+
179
+ ## 🔗 BibTeX
180
+
181
+ If you found this repository helpful, please cite our report:
182
+
183
+ ```bibtex
184
+ @misc{hunyuan3d22025tencent,
185
+ title={Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation},
186
+ author={Tencent Hunyuan3D Team},
187
+ year={2025},
188
+ eprint={2501.12202},
189
+ archivePrefix={arXiv},
190
+ primaryClass={cs.CV}
191
+ }
192
+
193
+ @misc{yang2024tencent,
194
+ title={Tencent Hunyuan3D-1.0: A Unified Framework for Text-to-3D and Image-to-3D Generation},
195
+ author={Tencent Hunyuan3D Team},
196
+ year={2024},
197
+ eprint={2411.02293},
198
+ archivePrefix={arXiv},
199
+ primaryClass={cs.CV}
200
+ }
201
+ ```
202
+
203
+ ## Community Resources
204
+
205
+ Thanks for the contributions of community members, here we have these great extensions of Hunyuan3D 2.0:
206
+
207
+ - [ComfyUI-Hunyuan3DWrapper](https://github.com/kijai/ComfyUI-Hunyuan3DWrapper)
208
+ - [Hunyuan3D-2-for-windows](https://github.com/sdbds/Hunyuan3D-2-for-windows)
209
+ - [📦 A bundle for running on Windows | 整合包](https://github.com/YanWenKun/Comfy3D-WinPortable/releases/tag/r8-hunyuan3d2)
210
+
211
+ ## Acknowledgements
212
+
213
+ We would like to thank the contributors to
214
+ the [DINOv2](https://github.com/facebookresearch/dinov2), [Stable Diffusion](https://github.com/Stability-AI/stablediffusion), [FLUX](https://github.com/black-forest-labs/flux), [diffusers](https://github.com/huggingface/diffusers)
215
+ and [HuggingFace](https://huggingface.co) repositories, for their open research and exploration.
216
+
217
+ ## Star History
218
+
219
+ <a href=""https://star-history.com/#Tencent/Hunyuan3D-2&Date"">
220
+ <picture>
221
+ <source media=""(prefers-color-scheme: dark)"" srcset=""https://api.star-history.com/svg?repos=Tencent/Hunyuan3D-2&type=Date&theme=dark"" />
222
+ <source media=""(prefers-color-scheme: light)"" srcset=""https://api.star-history.com/svg?repos=Tencent/Hunyuan3D-2&type=Date"" />
223
+ <img alt=""Star History Chart"" src=""https://api.star-history.com/svg?repos=Tencent/Hunyuan3D-2&type=Date"" />
224
+ </picture>
225
+ </a>","{""id"": ""tencent/Hunyuan3D-2"", ""author"": ""tencent"", ""sha"": ""34e28261f71c32975727be8db0eace439a280f82"", ""last_modified"": ""2025-04-10 14:55:56+00:00"", ""created_at"": ""2025-01-20 06:55:37+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 529128, ""downloads_all_time"": null, ""likes"": 1267, ""library_name"": ""hunyuan3d-2"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""hunyuan3d-2"", ""diffusers"", ""safetensors"", ""image-to-3d"", ""text-to-3d"", ""en"", ""zh"", ""arxiv:2501.12202"", ""arxiv:2411.02293"", ""license:other"", ""region:us""], ""pipeline_tag"": ""image-to-3d"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""language:\n- en\n- zh\nlibrary_name: hunyuan3d-2\nlicense: other\nlicense_name: tencent-hunyuan-community\nlicense_link: https://huggingface.co/tencent/Hunyuan3D-2/blob/main/LICENSE.txt\npipeline_tag: image-to-3d\ntags:\n- image-to-3d\n- text-to-3d"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='LICENSE', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='NOTICE', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='assets/demo.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='assets/images/arch.jpg', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='assets/images/e2e-1.gif', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='assets/images/e2e-2.gif', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='assets/images/system.jpg', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='assets/images/teaser.jpg', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-delight-v2-0/feature_extractor/preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-delight-v2-0/model_index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-delight-v2-0/scheduler/scheduler_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-delight-v2-0/text_encoder/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-delight-v2-0/text_encoder/model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-delight-v2-0/tokenizer/merges.txt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-delight-v2-0/tokenizer/special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-delight-v2-0/tokenizer/tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-delight-v2-0/tokenizer/vocab.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-delight-v2-0/unet/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-delight-v2-0/unet/diffusion_pytorch_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-delight-v2-0/vae/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-delight-v2-0/vae/diffusion_pytorch_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-dit-v2-0-fast/config.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-dit-v2-0-fast/model.fp16.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-dit-v2-0-fast/model.fp16.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-dit-v2-0-turbo/config.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-dit-v2-0-turbo/model.fp16.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-dit-v2-0-turbo/model.fp16.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-dit-v2-0/config.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-dit-v2-0/model.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-dit-v2-0/model.fp16.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-dit-v2-0/model.fp16.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-dit-v2-0/model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-dit-v2-0/model_fp16.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/feature_extractor/preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/image_encoder/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/image_encoder/model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/image_encoder/preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/model_index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/scheduler/scheduler_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/text_encoder/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/text_encoder/pytorch_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/tokenizer/merges.txt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/tokenizer/special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/tokenizer/tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/tokenizer/vocab.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/unet/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/unet/diffusion_pytorch_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/unet/diffusion_pytorch_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/unet/modules.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/vae/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0-turbo/vae/diffusion_pytorch_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/feature_extractor/preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/model_index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/scheduler/scheduler_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/text_encoder/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/text_encoder/pytorch_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/tokenizer/merges.txt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/tokenizer/special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/tokenizer/tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/tokenizer/vocab.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/unet/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/unet/diffusion_pytorch_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/unet/diffusion_pytorch_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/unet/modules.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/vae/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/vae/diffusion_pytorch_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-paint-v2-0/vae/diffusion_pytorch_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-vae-v2-0-turbo/config.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-vae-v2-0-turbo/model.fp16.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-vae-v2-0-turbo/model.fp16.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-vae-v2-0/config.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-vae-v2-0/model.fp16.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hunyuan3d-vae-v2-0/model.fp16.safetensors', size=None, blob_id=None, lfs=None)""], ""spaces"": [""tencent/Hunyuan3D-2"", ""tencent/Hunyuan3D-2mv"", ""tencent/Hunyuan3D-2mini-Turbo"", ""usenrame/mit-reality-hack-img-3d"", ""model2/advanceblur"", ""mubarak-alketbi/Hunyuan3D-2mini-Turbo"", ""Wkatir/Hunyuan3D-2"", ""mukaist/Hunyuan3D-2"", ""SharafeevRavil/test"", ""inoculatemedia/Hunyuan3D-2"", ""curryporkchop/TextTo3D_2.0"", ""Mahavaury2/consent_project"", ""Nymbo/Hunyuan3D-2"", ""ssbagpcm/Hunyuan3D-2"", ""MMD-Coder/Hunyuan3D-2.0"", ""davidvgilmore/hunyuan3d-custom"", ""davidvgilmore/hunyuan3d-lfs"", ""sizifart/siz3d"", ""NikoNovice/test2"", ""aimpowerment/Hunyuan3D-2"", ""rupert777/Hunyuan3D-2-cpu-test"", ""dawood/Hunyuan3D-2"", ""XtewaldX/Hunyuan3D-2"", ""sariyam/i-3d"", ""sariyam/t-3d"", ""Dhdb/Hunyuan3D-2"", ""syedMohib44/ditto-api"", ""DannyWoogagongtayafull/Hunyuan3D-2mini-Turbo"", ""mubarak-alketbi/Hunyuan3D-2"", ""icyleaf7/Hunyuan3D-2"", ""mohamedsobhi777/FramerComfy_basic11_8694404279"", ""mohamedsobhi777/FramerComfy_basic_2025_9272427525"", ""Rogerjs/Listto3d"", ""paceyai/Hunyuan3D-2mini-Turbo"", ""sasu1205/3d_tuning"", ""xinjjj/ImgRoboAssetGen"", ""shixiangbupt/Hunyuan3D-2"", ""VASANTHTHYU/Hunyuan3D-2"", ""mohamedsobhi777/FramerComfy_Basicflow_3191895845"", ""mohamedsobhi777/FramerComfy_FluxSchnell_9906338852"", ""millatmdd/Hunyuan3D-2_millat""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-04-10 14:55:56+00:00"", ""cardData"": ""language:\n- en\n- zh\nlibrary_name: hunyuan3d-2\nlicense: other\nlicense_name: tencent-hunyuan-community\nlicense_link: https://huggingface.co/tencent/Hunyuan3D-2/blob/main/LICENSE.txt\npipeline_tag: image-to-3d\ntags:\n- image-to-3d\n- text-to-3d"", ""transformersInfo"": null, ""_id"": ""678df3695dec6df8ec20e664"", ""modelId"": ""tencent/Hunyuan3D-2"", ""usedStorage"": 60813075231}",0,"https://huggingface.co/nomnom45465/Jane, https://huggingface.co/SergeySvichkar/TestModel, https://huggingface.co/Winrion/ebony, https://huggingface.co/everhard1/alpaca",4,,0,https://huggingface.co/calcuis/hy3d-gguf,1,,0,"SharafeevRavil/test, Wkatir/Hunyuan3D-2, huggingface/InferenceSupport/discussions/379, icyleaf7/Hunyuan3D-2, inoculatemedia/Hunyuan3D-2, model2/advanceblur, mohamedsobhi777/FramerComfy_basic11_8694404279, mohamedsobhi777/FramerComfy_basic_2025_9272427525, mubarak-alketbi/Hunyuan3D-2mini-Turbo, mukaist/Hunyuan3D-2, tencent/Hunyuan3D-2, tencent/Hunyuan3D-2mini-Turbo, tencent/Hunyuan3D-2mv",13
226
+ nomnom45465/Jane,"---
227
+ license: apache-2.0
228
+ datasets:
229
+ - bespokelabs/Bespoke-Stratos-17k
230
+ language:
231
+ - en
232
+ base_model:
233
+ - tencent/Hunyuan3D-2
234
+ new_version: tencent/Hunyuan3D-2
235
+ pipeline_tag: text-to-image
236
+ tags:
237
+ - not-for-all-audiences
238
+ ---
239
+ # Model Card for Model ID
240
+
241
+ <!-- Provide a quick summary of what the model is/does. -->
242
+
243
+ This modelcard aims to be a base template for new models. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/modelcard_template.md?plain=1).
244
+
245
+ ## Model Details
246
+
247
+ ### Model Description
248
+
249
+ <!-- Provide a longer summary of what this model is. -->
250
+
251
+
252
+
253
+ - **Developed by:** [More Information Needed]
254
+ - **Funded by [optional]:** [More Information Needed]
255
+ - **Shared by [optional]:** [More Information Needed]
256
+ - **Model type:** [More Information Needed]
257
+ - **Language(s) (NLP):** [More Information Needed]
258
+ - **License:** [More Information Needed]
259
+ - **Finetuned from model [optional]:** [More Information Needed]
260
+
261
+ ### Model Sources [optional]
262
+
263
+ <!-- Provide the basic links for the model. -->
264
+
265
+ - **Repository:** [More Information Needed]
266
+ - **Paper [optional]:** [More Information Needed]
267
+ - **Demo [optional]:** [More Information Needed]
268
+
269
+ ## Uses
270
+
271
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
272
+
273
+ ### Direct Use
274
+
275
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
276
+
277
+ [More Information Needed]
278
+
279
+ ### Downstream Use [optional]
280
+
281
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
282
+
283
+ [More Information Needed]
284
+
285
+ ### Out-of-Scope Use
286
+
287
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
288
+
289
+ [More Information Needed]
290
+
291
+ ## Bias, Risks, and Limitations
292
+
293
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
294
+
295
+ [More Information Needed]
296
+
297
+ ### Recommendations
298
+
299
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
300
+
301
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
302
+
303
+ ## How to Get Started with the Model
304
+
305
+ Use the code below to get started with the model.
306
+
307
+ [More Information Needed]
308
+
309
+ ## Training Details
310
+
311
+ ### Training Data
312
+
313
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
314
+
315
+ [More Information Needed]
316
+
317
+ ### Training Procedure
318
+
319
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
320
+
321
+ #### Preprocessing [optional]
322
+
323
+ [More Information Needed]
324
+
325
+
326
+ #### Training Hyperparameters
327
+
328
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
329
+
330
+ #### Speeds, Sizes, Times [optional]
331
+
332
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
333
+
334
+ [More Information Needed]
335
+
336
+ ## Evaluation
337
+
338
+ <!-- This section describes the evaluation protocols and provides the results. -->
339
+
340
+ ### Testing Data, Factors & Metrics
341
+
342
+ #### Testing Data
343
+
344
+ <!-- This should link to a Dataset Card if possible. -->
345
+
346
+ [More Information Needed]
347
+
348
+ #### Factors
349
+
350
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
351
+
352
+ [More Information Needed]
353
+
354
+ #### Metrics
355
+
356
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
357
+
358
+ [More Information Needed]
359
+
360
+ ### Results
361
+
362
+ [More Information Needed]
363
+
364
+ #### Summary
365
+
366
+
367
+
368
+ ## Model Examination [optional]
369
+
370
+ <!-- Relevant interpretability work for the model goes here -->
371
+
372
+ [More Information Needed]
373
+
374
+ ## Environmental Impact
375
+
376
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
377
+
378
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
379
+
380
+ - **Hardware Type:** [More Information Needed]
381
+ - **Hours used:** [More Information Needed]
382
+ - **Cloud Provider:** [More Information Needed]
383
+ - **Compute Region:** [More Information Needed]
384
+ - **Carbon Emitted:** [More Information Needed]
385
+
386
+ ## Technical Specifications [optional]
387
+
388
+ ### Model Architecture and Objective
389
+
390
+ [More Information Needed]
391
+
392
+ ### Compute Infrastructure
393
+
394
+ [More Information Needed]
395
+
396
+ #### Hardware
397
+
398
+ [More Information Needed]
399
+
400
+ #### Software
401
+
402
+ [More Information Needed]
403
+
404
+ ## Citation [optional]
405
+
406
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
407
+
408
+ **BibTeX:**
409
+
410
+ [More Information Needed]
411
+
412
+ **APA:**
413
+
414
+ [More Information Needed]
415
+
416
+ ## Glossary [optional]
417
+
418
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
419
+
420
+ [More Information Needed]
421
+
422
+ ## More Information [optional]
423
+
424
+ [More Information Needed]
425
+
426
+ ## Model Card Authors [optional]
427
+
428
+ [More Information Needed]
429
+
430
+ ## Model Card Contact
431
+
432
+ [More Information Needed]","{""id"": ""nomnom45465/Jane"", ""author"": ""nomnom45465"", ""sha"": ""84556b03709955c3e9d9d81c37ea077b1d6a9d4b"", ""last_modified"": ""2025-02-17 03:43:35+00:00"", ""created_at"": ""2025-02-17 03:41:58+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 1, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""not-for-all-audiences"", ""text-to-image"", ""en"", ""dataset:bespokelabs/Bespoke-Stratos-17k"", ""arxiv:1910.09700"", ""base_model:tencent/Hunyuan3D-2"", ""base_model:finetune:tencent/Hunyuan3D-2"", ""license:apache-2.0"", ""region:us""], ""pipeline_tag"": ""text-to-image"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- tencent/Hunyuan3D-2\ndatasets:\n- bespokelabs/Bespoke-Stratos-17k\nlanguage:\n- en\nlicense: apache-2.0\npipeline_tag: text-to-image\ntags:\n- not-for-all-audiences\nnew_version: tencent/Hunyuan3D-2"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-17 03:43:35+00:00"", ""cardData"": ""base_model:\n- tencent/Hunyuan3D-2\ndatasets:\n- bespokelabs/Bespoke-Stratos-17k\nlanguage:\n- en\nlicense: apache-2.0\npipeline_tag: text-to-image\ntags:\n- not-for-all-audiences\nnew_version: tencent/Hunyuan3D-2"", ""transformersInfo"": null, ""_id"": ""67b2b0067e1d6634494ce627"", ""modelId"": ""nomnom45465/Jane"", ""usedStorage"": 0}",1,,0,,0,,0,,0,,0
433
+ SergeySvichkar/TestModel,"---
434
+ language:
435
+ - en
436
+ base_model:
437
+ - tencent/Hunyuan3D-2
438
+ ---","{""id"": ""SergeySvichkar/TestModel"", ""author"": ""SergeySvichkar"", ""sha"": ""275d5c2e800c5f13d9fb198d1ebd6d901cca0c0f"", ""last_modified"": ""2025-01-27 00:57:02+00:00"", ""created_at"": ""2025-01-27 00:55:50+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""en"", ""base_model:tencent/Hunyuan3D-2"", ""base_model:finetune:tencent/Hunyuan3D-2"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- tencent/Hunyuan3D-2\nlanguage:\n- en"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-01-27 00:57:02+00:00"", ""cardData"": ""base_model:\n- tencent/Hunyuan3D-2\nlanguage:\n- en"", ""transformersInfo"": null, ""_id"": ""6796d996d4afc6fb1c8090f4"", ""modelId"": ""SergeySvichkar/TestModel"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=SergeySvichkar/TestModel&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BSergeySvichkar%2FTestModel%5D(%2FSergeySvichkar%2FTestModel)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
439
+ Winrion/ebony,"---
440
+ license: apache-2.0
441
+ base_model:
442
+ - tencent/Hunyuan3D-2
443
+ pipeline_tag: feature-extraction
444
+ ---","{""id"": ""Winrion/ebony"", ""author"": ""Winrion"", ""sha"": ""52d0886a88bc459ecdba379f01971ea6b848e84a"", ""last_modified"": ""2025-02-05 05:12:33+00:00"", ""created_at"": ""2025-02-05 05:11:45+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""feature-extraction"", ""base_model:tencent/Hunyuan3D-2"", ""base_model:finetune:tencent/Hunyuan3D-2"", ""license:apache-2.0"", ""region:us""], ""pipeline_tag"": ""feature-extraction"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- tencent/Hunyuan3D-2\nlicense: apache-2.0\npipeline_tag: feature-extraction"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-05 05:12:33+00:00"", ""cardData"": ""base_model:\n- tencent/Hunyuan3D-2\nlicense: apache-2.0\npipeline_tag: feature-extraction"", ""transformersInfo"": null, ""_id"": ""67a2f3114fdf4d91876c0fbe"", ""modelId"": ""Winrion/ebony"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=Winrion/ebony&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BWinrion%2Febony%5D(%2FWinrion%2Febony)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
445
+ everhard1/alpaca,"---
446
+ license: apache-2.0
447
+ datasets:
448
+ - fka/awesome-chatgpt-prompts
449
+ language:
450
+ - en
451
+ metrics:
452
+ - accuracy
453
+ base_model:
454
+ - tencent/Hunyuan3D-2
455
+ new_version: deepseek-ai/DeepSeek-R1
456
+ library_name: flair
457
+ tags:
458
+ - art
459
+ ---","{""id"": ""everhard1/alpaca"", ""author"": ""everhard1"", ""sha"": ""e54bd3b300ddecfca640fa003b17d07243b434ab"", ""last_modified"": ""2025-02-11 21:20:53+00:00"", ""created_at"": ""2025-02-11 21:19:57+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""flair"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""flair"", ""art"", ""en"", ""dataset:fka/awesome-chatgpt-prompts"", ""base_model:tencent/Hunyuan3D-2"", ""base_model:finetune:tencent/Hunyuan3D-2"", ""license:apache-2.0"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- tencent/Hunyuan3D-2\ndatasets:\n- fka/awesome-chatgpt-prompts\nlanguage:\n- en\nlibrary_name: flair\nlicense: apache-2.0\nmetrics:\n- accuracy\ntags:\n- art\nnew_version: deepseek-ai/DeepSeek-R1"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-11 21:20:53+00:00"", ""cardData"": ""base_model:\n- tencent/Hunyuan3D-2\ndatasets:\n- fka/awesome-chatgpt-prompts\nlanguage:\n- en\nlibrary_name: flair\nlicense: apache-2.0\nmetrics:\n- accuracy\ntags:\n- art\nnew_version: deepseek-ai/DeepSeek-R1"", ""transformersInfo"": null, ""_id"": ""67abbefdb8f6af16085abd42"", ""modelId"": ""everhard1/alpaca"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=everhard1/alpaca&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Beverhard1%2Falpaca%5D(%2Feverhard1%2Falpaca)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
Idefics3-8B-Llama3_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv ADDED
@@ -0,0 +1,607 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ HuggingFaceM4/Idefics3-8B-Llama3,"---
3
+ license: apache-2.0
4
+ datasets:
5
+ - HuggingFaceM4/OBELICS
6
+ - HuggingFaceM4/the_cauldron
7
+ - HuggingFaceM4/Docmatix
8
+ - HuggingFaceM4/WebSight
9
+ language:
10
+ - en
11
+ tags:
12
+ - multimodal
13
+ - vision
14
+ - image-text-to-text
15
+ library_name: transformers
16
+ ---
17
+
18
+ <p align=""center"">
19
+ <img src=""https://huggingface.co/HuggingFaceM4/idefics-80b/resolve/main/assets/IDEFICS.png"" alt=""Idefics-Obelics logo"" width=""200"" height=""100"">
20
+ </p>
21
+
22
+ **Transformers version**: >4.46.
23
+
24
+ # Idefics3
25
+
26
+ Idefics3 is an open multimodal model that accepts arbitrary sequences of image and text inputs and produces text outputs. The model can answer questions about images, describe visual content, create stories grounded on multiple images, or simply behave as a pure language model without visual inputs. It improves upon [Idefics1](https://huggingface.co/HuggingFaceM4/idefics-80b-instruct) and [Idefics2](https://huggingface.co/HuggingFaceM4/idefics2-8b), significantly enhancing capabilities around OCR, document understanding and visual reasoning.
27
+
28
+ We release the checkpoints under the Apache 2.0.
29
+
30
+ # Model Summary
31
+
32
+ - **Developed by:** Hugging Face
33
+ - **Model type:** Multi-modal model (image+text)
34
+ - **Language(s) (NLP):** en
35
+ - **License:** Apache 2.0
36
+ - **Parent Models:** [google/siglip-so400m-patch14-384](https://huggingface.co/google/siglip-so400m-patch14-384) and [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct)
37
+ - **Resources for more information:**
38
+ - Idefics1 paper: [OBELICS: An Open Web-Scale Filtered Dataset of Interleaved Image-Text Documents
39
+ ](https://huggingface.co/papers/2306.16527)
40
+ - Idefics2 paper: [What matters when building vision-language models?
41
+ ](https://huggingface.co/papers/2405.02246)
42
+ - Idefics3 paper: [Building and better understanding vision-language models: insights and future directions](https://huggingface.co/papers/2408.12637)
43
+
44
+ # Uses
45
+
46
+ `Idefics3-8B` can be used to perform inference on multimodal (image + text) tasks in which the input is composed of a text query along with one (or multiple) image(s). Text and images can be arbitrarily interleaved. That includes image captioning, visual question answering, etc. These model does not support image generation.
47
+
48
+ The post-training of Idefics3-8B involves only a supervised fine-tuning stage, without RLHF alignment. As a result, the model may produce short answers or require prompt iterations to fully address the user's request. Adding a prefix to the assistant's response, such as ""Let's fix this step by step"" has been found to effectively influence the generated output.
49
+
50
+ To fine-tune `Idefics3-8B` on a specific task, we provide a [fine-tuning tutorial](https://github.com/merveenoyan/smol-vision/blob/main/Idefics_FT.ipynb).
51
+ Other resources for the fine-tuning of Idefics2 (can easily be adapted to Idefics3):
52
+ - With the [TRL library](https://github.com/huggingface/trl): [Script](https://gist.github.com/edbeeching/228652fc6c2b29a1641be5a5778223cb)
53
+ - With the [Hugging Face Trainer](https://huggingface.co/docs/transformers/main/en/main_classes/trainer#api-reference%20][%20transformers.Trainer): [Tutorial notebook](https://colab.research.google.com/drive/1NtcTgRbSBKN7pYD3Vdx1j9m8pt3fhFDB?usp=sharing)
54
+
55
+
56
+ # Technical summary
57
+
58
+ Idefics3 demonstrates a great improvement over Idefics2, especially in document understanding tasks. It serves as a strong foundation for various use-case specific fine-tunings.
59
+
60
+ | Model | MMMU <br>(val) | MathVista <br>(test) | MMStar <br>(val) | DocVQA <br>(test) | TextVQA <br>(val) |
61
+ |:---------------:|:----------------:|:----------------------:|:-------------------:|:--------------------:|:-----------------:|
62
+ | **Idefics2-8B** | 45.2 | 52.2 | 49.5 | 74.0 | 73.0 |
63
+ | **Idefics3-8B** | 46.6 | 58.4 | 55.9 | 87.7 | 74.9 |
64
+
65
+
66
+ **Idefics3 introduces several changes compared to Idefics2:**
67
+ - We use 169 visual tokens to encode a image of size 364x364. Each image is divided into several sub images of sizes at most 364x364, which are then encoded separately.
68
+ - For the fine-tuning datasets, we have extended [The Cauldron](https://huggingface.co/datasets/HuggingFaceM4/the_cauldron) and added several datasets, including [Docmatix](HuggingFaceM4/Docmatix). We will push soon these datasets to the same repo of The Cauldron (TODO).
69
+
70
+ More details about the training of the model is available in our [technical report](https://huggingface.co/papers/2408.12637).
71
+
72
+
73
+ # How to Get Started
74
+
75
+ This section shows snippets of code for generation for `Idefics3-8B`.
76
+
77
+ ```python
78
+ import requests
79
+ import torch
80
+ from PIL import Image
81
+ from io import BytesIO
82
+
83
+ from transformers import AutoProcessor, AutoModelForVision2Seq
84
+ from transformers.image_utils import load_image
85
+
86
+ DEVICE = ""cuda:0""
87
+
88
+ # Note that passing the image urls (instead of the actual pil images) to the processor is also possible
89
+ image1 = load_image(""https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"")
90
+ image2 = load_image(""https://cdn.britannica.com/59/94459-050-DBA42467/Skyline-Chicago.jpg"")
91
+ image3 = load_image(""https://cdn.britannica.com/68/170868-050-8DDE8263/Golden-Gate-Bridge-San-Francisco.jpg"")
92
+
93
+ processor = AutoProcessor.from_pretrained(""HuggingFaceM4/Idefics3-8B-Llama3"")
94
+ model = AutoModelForVision2Seq.from_pretrained(
95
+ ""HuggingFaceM4/Idefics3-8B-Llama3"", torch_dtype=torch.bfloat16
96
+ ).to(DEVICE)
97
+
98
+ # Create inputs
99
+ messages = [
100
+ {
101
+ ""role"": ""user"",
102
+ ""content"": [
103
+ {""type"": ""image""},
104
+ {""type"": ""text"", ""text"": ""What do we see in this image?""},
105
+ ]
106
+ },
107
+ {
108
+ ""role"": ""assistant"",
109
+ ""content"": [
110
+ {""type"": ""text"", ""text"": ""In this image, we can see the city of New York, and more specifically the Statue of Liberty.""},
111
+ ]
112
+ },
113
+ {
114
+ ""role"": ""user"",
115
+ ""content"": [
116
+ {""type"": ""image""},
117
+ {""type"": ""text"", ""text"": ""And how about this image?""},
118
+ ]
119
+ },
120
+ ]
121
+ prompt = processor.apply_chat_template(messages, add_generation_prompt=True)
122
+ inputs = processor(text=prompt, images=[image1, image2], return_tensors=""pt"")
123
+ inputs = {k: v.to(DEVICE) for k, v in inputs.items()}
124
+
125
+
126
+ # Generate
127
+ generated_ids = model.generate(**inputs, max_new_tokens=500)
128
+ generated_texts = processor.batch_decode(generated_ids, skip_special_tokens=True)
129
+
130
+ print(generated_texts)
131
+ ```
132
+
133
+ </details>
134
+
135
+ **Text generation inference**
136
+
137
+ TODO.
138
+
139
+ # Model optimizations
140
+
141
+ If your GPU allows, we first recommend loading (and running inference) in half precision (`torch.float16` or `torch.bfloat16`).
142
+
143
+ ```diff
144
+ model = AutoModelForVision2Seq.from_pretrained(
145
+ ""HuggingFaceM4/Idefics3-8B-Llama3"",
146
+ + torch_dtype=torch.bfloat16,
147
+ ).to(DEVICE)
148
+ ```
149
+
150
+ **Vision encoder efficiency**
151
+
152
+ You can choose the default resolution the images will be rescaled to by adding `size= {""longest_edge"": N*364}` when initializing the processor (`AutoProcessor.from_pretrained`), with `N` your desired value.
153
+ `N=4` works best in practice (this is the default value), but for very large images, it could be interesting to pass `N=5`.
154
+ This will have an impact on the number of visual tokens passed to the language model.
155
+ If you are GPU-memory-constrained, you can decrease `N`, and choose for example `N=3` or `N=2`, especially for low resolution images.
156
+
157
+ **Using Flash-attention 2 to speed up generation**
158
+
159
+ <details><summary>Click to expand.</summary>
160
+
161
+ First, make sure to install `flash-attn`. Refer to the [original repository of Flash Attention](https://github.com/Dao-AILab/flash-attention) for the package installation. Simply change the snippet above with:
162
+
163
+ ```diff
164
+ model = AutoModelForVision2Seq.from_pretrained(
165
+ ""HuggingFaceM4/Idefics3-8B-Llama3"",
166
+ + torch_dtype=torch.bfloat16,
167
+ + _attn_implementation=""flash_attention_2"",
168
+ ).to(DEVICE)
169
+ ```
170
+
171
+ </details>
172
+
173
+
174
+ # Misuse and Out-of-scope use
175
+
176
+ Using the model in [high-stakes](https://huggingface.co/bigscience/bloom/blob/main/README.md#glossary-and-calculations) settings is out of scope for this model. The model is not designed for [critical decisions](https://huggingface.co/bigscience/bloom/blob/main/README.md#glossary-and-calculations) nor uses with any material consequences on an individual's livelihood or wellbeing. The model outputs content that appears factual but may not be correct. Out-of-scope uses include:
177
+ - Usage for evaluating or scoring individuals, such as for employment, education, or credit
178
+ - Applying the model for critical automatic decisions, generating factual content, creating reliable summaries, or generating predictions that must be correct
179
+
180
+ Intentionally using the model for harm, violating [human rights](https://huggingface.co/bigscience/bloom/blob/main/README.md#glossary-and-calculations), or other kinds of malicious activities, is a misuse of this model. This includes:
181
+ - Spam generation
182
+ - Disinformation and influence operations
183
+ - Disparagement and defamation
184
+ - Harassment and abuse
185
+ - [Deception](https://huggingface.co/bigscience/bloom/blob/main/README.md#glossary-and-calculations)
186
+ - Unconsented impersonation and imitation
187
+ - Unconsented surveillance
188
+
189
+
190
+ # License
191
+
192
+ The model is built on top of two pre-trained models: [google/siglip-so400m-patch14-384](https://huggingface.co/google/siglip-so400m-patch14-384) and [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct). We release the Idefics3 checkpoints under the Apache 2.0 license.
193
+
194
+
195
+ # Citation
196
+
197
+ **BibTeX:**
198
+
199
+ ```bibtex
200
+ @misc{laurençon2024building,
201
+ title={Building and better understanding vision-language models: insights and future directions.},
202
+ author={Hugo Laurençon and Andrés Marafioti and Victor Sanh and Léo Tronchon},
203
+ year={2024},
204
+ eprint={2408.12637},
205
+ archivePrefix={arXiv},
206
+ primaryClass={cs.CV}
207
+ }
208
+ ```
209
+
210
+
211
+ # Acknowledgements
212
+
213
+ We thank @andito and @amyeroberts for helping on the integration in Transformers.","{""id"": ""HuggingFaceM4/Idefics3-8B-Llama3"", ""author"": ""HuggingFaceM4"", ""sha"": ""fddb4ff79181e55a994674777e06cd5456ce3dc3"", ""last_modified"": ""2024-12-02 09:35:34+00:00"", ""created_at"": ""2024-08-05 16:12:33+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 47806, ""downloads_all_time"": null, ""likes"": 276, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""safetensors"", ""idefics3"", ""image-text-to-text"", ""multimodal"", ""vision"", ""conversational"", ""en"", ""dataset:HuggingFaceM4/OBELICS"", ""dataset:HuggingFaceM4/the_cauldron"", ""dataset:HuggingFaceM4/Docmatix"", ""dataset:HuggingFaceM4/WebSight"", ""arxiv:2306.16527"", ""arxiv:2405.02246"", ""arxiv:2408.12637"", ""license:apache-2.0"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""image-text-to-text"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""datasets:\n- HuggingFaceM4/OBELICS\n- HuggingFaceM4/the_cauldron\n- HuggingFaceM4/Docmatix\n- HuggingFaceM4/WebSight\nlanguage:\n- en\nlibrary_name: transformers\nlicense: apache-2.0\ntags:\n- multimodal\n- vision\n- image-text-to-text"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": null, ""config"": {""architectures"": [""Idefics3ForConditionalGeneration""], ""model_type"": ""idefics3"", ""processor_config"": {""chat_template"": ""<|begin_of_text|>{% for message in messages %}{{message['role'].capitalize()}}{% if message['content'][0]['type'] == 'image' %}{{':'}}{% else %}{{': '}}{% endif %}{% for line in message['content'] %}{% if line['type'] == 'text' %}{{line['text']}}{% elif line['type'] == 'image' %}{{ '<image>' }}{% endif %}{% endfor %}<end_of_utterance>\n{% endfor %}{% if add_generation_prompt %}{{ 'Assistant:' }}{% endif %}""}, ""tokenizer_config"": {""bos_token"": ""<|begin_of_text|>"", ""chat_template"": ""{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}"", ""eos_token"": ""<|end_of_text|>"", ""pad_token"": ""<|reserved_special_token_0|>""}}, ""transformers_info"": {""auto_model"": ""AutoModelForImageTextToText"", ""custom_class"": null, ""pipeline_tag"": ""image-text-to-text"", ""processor"": ""AutoProcessor""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='chat_template.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00001-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00002-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00003-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00004-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [""HuggingFaceM4/idefics3"", ""TIGER-Lab/MEGA-Bench"", ""eltorio/IDEFICS3_ROCO"", ""AdrienB134/rag_colpali_idefics3"", ""hexgrad/IDEFICS3_ROCO_ZeroGPU"", ""arad1367/Marketing_Vision_HuggingFaceM4_idefics3"", ""acecalisto3/IDEfix"", ""d-delaurier/Judge-vLLM"", ""emoud/IDEFICS3_ROCO"", ""mcouaillac/IDEFICS3_ROCO_ZeroGPU"", ""awacke1/Leaderboard-Deepseek-Gemini-Grok-GPT-Qwen"", ""Zaherrr/KG_transform"", ""jkorstad/idefics3"", ""fatima3597/AI-Podcast-Creator"", ""jlecocq/radiology-test"", ""cmaire/IDEFICS3_ROCO_ZeroGPU"", ""cmaire/IDEFICS3_ROCO""], ""safetensors"": {""parameters"": {""BF16"": 8462086384}, ""total"": 8462086384}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-12-02 09:35:34+00:00"", ""cardData"": ""datasets:\n- HuggingFaceM4/OBELICS\n- HuggingFaceM4/the_cauldron\n- HuggingFaceM4/Docmatix\n- HuggingFaceM4/WebSight\nlanguage:\n- en\nlibrary_name: transformers\nlicense: apache-2.0\ntags:\n- multimodal\n- vision\n- image-text-to-text"", ""transformersInfo"": {""auto_model"": ""AutoModelForImageTextToText"", ""custom_class"": null, ""pipeline_tag"": ""image-text-to-text"", ""processor"": ""AutoProcessor""}, ""_id"": ""66b0f9f15cb4654fd1d665bf"", ""modelId"": ""HuggingFaceM4/Idefics3-8B-Llama3"", ""usedStorage"": 16924267272}",0,"https://huggingface.co/Minthy/ToriiGate-v0.3, https://huggingface.co/nectec/Pathumma-llm-vision-1.0.0, https://huggingface.co/Mantis-VL/mantis-8b-idefics3_16384, https://huggingface.co/Mantis-VL/mantis-8b-idefics3-pure_16384, https://huggingface.co/Leeyuyu/idefics3-llama-thyroid, https://huggingface.co/Clark12/POS1, https://huggingface.co/Minthy/Torii_Gate_v0.1_alpha, https://huggingface.co/slezki/assistant_mage, https://huggingface.co/Minthy/ToriiGate-v0.2",9,"https://huggingface.co/eltorio/IDEFICS3_ROCOv2, https://huggingface.co/joris-sense/idefics3-llama-vqav2, https://huggingface.co/fsommers/idefics3-llama-vqav2_1, https://huggingface.co/Maverick17/idefics3-llama-gui-dense-descriptions, https://huggingface.co/justinkarlin/idefics3-llama-vqav2, https://huggingface.co/justinkarlin/idefics3-qlora-faces, https://huggingface.co/justinkarlin/idefics3-qlora-faces2, https://huggingface.co/justinkarlin/idefics3-qlora-faces3, https://huggingface.co/justinkarlin/idefics3-qlora-faces4, https://huggingface.co/Erland/idefics3-llama-ai701, https://huggingface.co/eltorio/IDEFICS3_ROCO, https://huggingface.co/eltorio/IDEFICS3_medical_instruct, https://huggingface.co/justinkarlin/idefics3-qlora-faces5, https://huggingface.co/justinkarlin/idefics3-qlora-faces6, https://huggingface.co/kurama270296/qlora-idefics3-test, https://huggingface.co/Tchalla12/idefics3-8b-qlora",16,"https://huggingface.co/2dameneko/Idefics3-8B-Llama3-nf4, https://huggingface.co/leon-se/Idefics3-8B-Llama3-bnb_nf4, https://huggingface.co/leon-se/Idefics3-8B-Llama3-FP8-Dynamic",3,,0,"AdrienB134/rag_colpali_idefics3, HuggingFaceM4/idefics3, TIGER-Lab/MEGA-Bench, acecalisto3/IDEfix, arad1367/Marketing_Vision_HuggingFaceM4_idefics3, awacke1/Leaderboard-Deepseek-Gemini-Grok-GPT-Qwen, d-delaurier/Judge-vLLM, eltorio/IDEFICS3_ROCO, emoud/IDEFICS3_ROCO, hexgrad/IDEFICS3_ROCO_ZeroGPU, huggingface/InferenceSupport/discussions/new?title=HuggingFaceM4/Idefics3-8B-Llama3&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BHuggingFaceM4%2FIdefics3-8B-Llama3%5D(%2FHuggingFaceM4%2FIdefics3-8B-Llama3)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, jkorstad/idefics3, mcouaillac/IDEFICS3_ROCO_ZeroGPU",13
214
+ https://huggingface.co/Minthy/ToriiGate-v0.3,N/A,N/A,1,,0,,0,,0,,0,,0
215
+ https://huggingface.co/nectec/Pathumma-llm-vision-1.0.0,N/A,N/A,1,,0,,0,,0,,0,,0
216
+ Mantis-VL/mantis-8b-idefics3_16384,"---
217
+ library_name: transformers
218
+ license: apache-2.0
219
+ base_model: HuggingFaceM4/Idefics3-8B-Llama3
220
+ tags:
221
+ - generated_from_trainer
222
+ model-index:
223
+ - name: mantis-8b-idefics3_16384
224
+ results: []
225
+ ---
226
+
227
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
228
+ should probably proofread and complete it, then remove this comment. -->
229
+
230
+ # mantis-8b-idefics3_16384
231
+
232
+ This model is a fine-tuned version of [HuggingFaceM4/Idefics3-8B-Llama3](https://huggingface.co/HuggingFaceM4/Idefics3-8B-Llama3) on an unknown dataset.
233
+
234
+ ## Model description
235
+
236
+ More information needed
237
+
238
+ ## Intended uses & limitations
239
+
240
+ More information needed
241
+
242
+ ## Training and evaluation data
243
+
244
+ More information needed
245
+
246
+ ## Training procedure
247
+
248
+ ### Training hyperparameters
249
+
250
+ The following hyperparameters were used during training:
251
+ - learning_rate: 5e-06
252
+ - train_batch_size: 1
253
+ - eval_batch_size: 1
254
+ - seed: 42
255
+ - distributed_type: multi-GPU
256
+ - num_devices: 16
257
+ - gradient_accumulation_steps: 8
258
+ - total_train_batch_size: 128
259
+ - total_eval_batch_size: 16
260
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
261
+ - lr_scheduler_type: cosine
262
+ - lr_scheduler_warmup_ratio: 0.03
263
+ - num_epochs: 1.0
264
+
265
+ ### Training results
266
+
267
+
268
+
269
+ ### Framework versions
270
+
271
+ - Transformers 4.45.0.dev0
272
+ - Pytorch 2.3.0+cu121
273
+ - Datasets 2.18.0
274
+ - Tokenizers 0.19.1
275
+ ","{""id"": ""Mantis-VL/mantis-8b-idefics3_16384"", ""author"": ""Mantis-VL"", ""sha"": ""ca959abe0d86e62ac272619aea4e49b57fbe7ad0"", ""last_modified"": ""2024-09-03 21:57:21+00:00"", ""created_at"": ""2024-09-02 09:40:24+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 2, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""safetensors"", ""idefics3"", ""image-text-to-text"", ""generated_from_trainer"", ""conversational"", ""base_model:HuggingFaceM4/Idefics3-8B-Llama3"", ""base_model:finetune:HuggingFaceM4/Idefics3-8B-Llama3"", ""license:apache-2.0"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""image-text-to-text"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: HuggingFaceM4/Idefics3-8B-Llama3\nlibrary_name: transformers\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: mantis-8b-idefics3_16384\n results: []"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": [{""name"": ""mantis-8b-idefics3_16384"", ""results"": []}], ""config"": {""architectures"": [""Idefics3ForConditionalGeneration""], ""model_type"": ""idefics3"", ""processor_config"": {""chat_template"": ""<|begin_of_text|>{% for message in messages %}{{message['role'].capitalize()}}{% if message['content'][0]['type'] == 'image' %}{{':'}}{% else %}{{': '}}{% endif %}{% for line in message['content'] %}{% if line['type'] == 'text' %}{{line['text']}}{% elif line['type'] == 'image' %}{{ '<image>' }}{% endif %}{% endfor %}<end_of_utterance>\n{% endfor %}{% if add_generation_prompt %}{{ 'Assistant:' }}{% endif %}""}, ""tokenizer_config"": {""bos_token"": ""<|begin_of_text|>"", ""chat_template"": ""{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}"", ""eos_token"": ""<|end_of_text|>"", ""pad_token"": ""<|reserved_special_token_0|>""}}, ""transformers_info"": {""auto_model"": ""AutoModelForImageTextToText"", ""custom_class"": null, ""pipeline_tag"": ""image-text-to-text"", ""processor"": ""AutoProcessor""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='chat_template.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00001-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00002-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00003-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00004-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='training_args.bin', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": {""parameters"": {""BF16"": 8462086384}, ""total"": 8462086384}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-09-03 21:57:21+00:00"", ""cardData"": ""base_model: HuggingFaceM4/Idefics3-8B-Llama3\nlibrary_name: transformers\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: mantis-8b-idefics3_16384\n results: []"", ""transformersInfo"": {""auto_model"": ""AutoModelForImageTextToText"", ""custom_class"": null, ""pipeline_tag"": ""image-text-to-text"", ""processor"": ""AutoProcessor""}, ""_id"": ""66d588088a438492b0d0c548"", ""modelId"": ""Mantis-VL/mantis-8b-idefics3_16384"", ""usedStorage"": 226940140648}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=Mantis-VL/mantis-8b-idefics3_16384&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BMantis-VL%2Fmantis-8b-idefics3_16384%5D(%2FMantis-VL%2Fmantis-8b-idefics3_16384)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
276
+ Mantis-VL/mantis-8b-idefics3-pure_16384,"---
277
+ library_name: transformers
278
+ license: apache-2.0
279
+ base_model: HuggingFaceM4/Idefics3-8B-Llama3
280
+ tags:
281
+ - generated_from_trainer
282
+ model-index:
283
+ - name: mantis-8b-idefics3-pure_16384
284
+ results: []
285
+ ---
286
+
287
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
288
+ should probably proofread and complete it, then remove this comment. -->
289
+
290
+ # mantis-8b-idefics3-pure_16384
291
+
292
+ This model is a fine-tuned version of [HuggingFaceM4/Idefics3-8B-Llama3](https://huggingface.co/HuggingFaceM4/Idefics3-8B-Llama3) on an unknown dataset.
293
+
294
+ ## Model description
295
+
296
+ More information needed
297
+
298
+ ## Intended uses & limitations
299
+
300
+ More information needed
301
+
302
+ ## Training and evaluation data
303
+
304
+ More information needed
305
+
306
+ ## Training procedure
307
+
308
+ ### Training hyperparameters
309
+
310
+ The following hyperparameters were used during training:
311
+ - learning_rate: 5e-06
312
+ - train_batch_size: 1
313
+ - eval_batch_size: 1
314
+ - seed: 42
315
+ - distributed_type: multi-GPU
316
+ - num_devices: 16
317
+ - gradient_accumulation_steps: 8
318
+ - total_train_batch_size: 128
319
+ - total_eval_batch_size: 16
320
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
321
+ - lr_scheduler_type: cosine
322
+ - lr_scheduler_warmup_ratio: 0.03
323
+ - num_epochs: 1.0
324
+
325
+ ### Training results
326
+
327
+
328
+
329
+ ### Framework versions
330
+
331
+ - Transformers 4.45.0.dev0
332
+ - Pytorch 2.3.0+cu121
333
+ - Datasets 2.18.0
334
+ - Tokenizers 0.19.1
335
+ ","{""id"": ""Mantis-VL/mantis-8b-idefics3-pure_16384"", ""author"": ""Mantis-VL"", ""sha"": ""9d58e6dfccbf1f8cce51167119ff3a9d9f124b82"", ""last_modified"": ""2024-09-05 15:51:19+00:00"", ""created_at"": ""2024-09-04 10:01:16+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 2, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""safetensors"", ""idefics3"", ""image-text-to-text"", ""generated_from_trainer"", ""conversational"", ""base_model:HuggingFaceM4/Idefics3-8B-Llama3"", ""base_model:finetune:HuggingFaceM4/Idefics3-8B-Llama3"", ""license:apache-2.0"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""image-text-to-text"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: HuggingFaceM4/Idefics3-8B-Llama3\nlibrary_name: transformers\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: mantis-8b-idefics3-pure_16384\n results: []"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": [{""name"": ""mantis-8b-idefics3-pure_16384"", ""results"": []}], ""config"": {""architectures"": [""Idefics3ForConditionalGeneration""], ""model_type"": ""idefics3"", ""processor_config"": {""chat_template"": ""<|begin_of_text|>{% for message in messages %}{{message['role'].capitalize()}}{% if message['content'][0]['type'] == 'image' %}{{':'}}{% else %}{{': '}}{% endif %}{% for line in message['content'] %}{% if line['type'] == 'text' %}{{line['text']}}{% elif line['type'] == 'image' %}{{ '<image>' }}{% endif %}{% endfor %}<end_of_utterance>\n{% endfor %}{% if add_generation_prompt %}{{ 'Assistant:' }}{% endif %}""}, ""tokenizer_config"": {""bos_token"": ""<|begin_of_text|>"", ""chat_template"": ""{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}"", ""eos_token"": ""<|end_of_text|>"", ""pad_token"": ""<|reserved_special_token_0|>""}}, ""transformers_info"": {""auto_model"": ""AutoModelForImageTextToText"", ""custom_class"": null, ""pipeline_tag"": ""image-text-to-text"", ""processor"": ""AutoProcessor""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='chat_template.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00001-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00002-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00003-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00004-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='training_args.bin', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": {""parameters"": {""BF16"": 8462086384}, ""total"": 8462086384}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-09-05 15:51:19+00:00"", ""cardData"": ""base_model: HuggingFaceM4/Idefics3-8B-Llama3\nlibrary_name: transformers\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: mantis-8b-idefics3-pure_16384\n results: []"", ""transformersInfo"": {""auto_model"": ""AutoModelForImageTextToText"", ""custom_class"": null, ""pipeline_tag"": ""image-text-to-text"", ""processor"": ""AutoProcessor""}, ""_id"": ""66d82fec38b749dea1e9cc07"", ""modelId"": ""Mantis-VL/mantis-8b-idefics3-pure_16384"", ""usedStorage"": 203091213784}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=Mantis-VL/mantis-8b-idefics3-pure_16384&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BMantis-VL%2Fmantis-8b-idefics3-pure_16384%5D(%2FMantis-VL%2Fmantis-8b-idefics3-pure_16384)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
336
+ Leeyuyu/idefics3-llama-thyroid,"---
337
+ license: apache-2.0
338
+ base_model: HuggingFaceM4/Idefics3-8B-Llama3
339
+ tags:
340
+ - generated_from_trainer
341
+ model-index:
342
+ - name: idefics3-llama-thyroid
343
+ results: []
344
+ ---
345
+
346
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
347
+ should probably proofread and complete it, then remove this comment. -->
348
+
349
+ # idefics3-llama-thyroid
350
+
351
+ This model is a fine-tuned version of [HuggingFaceM4/Idefics3-8B-Llama3](https://huggingface.co/HuggingFaceM4/Idefics3-8B-Llama3) on an unknown dataset.
352
+
353
+ ## Model description
354
+
355
+ More information needed
356
+
357
+ ## Intended uses & limitations
358
+
359
+ More information needed
360
+
361
+ ## Training and evaluation data
362
+
363
+ More information needed
364
+
365
+ ## Training procedure
366
+
367
+ ### Training hyperparameters
368
+
369
+ The following hyperparameters were used during training:
370
+ - learning_rate: 0.0001
371
+ - train_batch_size: 1
372
+ - eval_batch_size: 8
373
+ - seed: 42
374
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
375
+ - lr_scheduler_type: linear
376
+ - lr_scheduler_warmup_steps: 50
377
+ - num_epochs: 3
378
+
379
+ ### Training results
380
+
381
+
382
+
383
+ ### Framework versions
384
+
385
+ - Transformers 4.44.0.dev0
386
+ - Pytorch 2.1.0+cu118
387
+ - Datasets 2.21.0
388
+ - Tokenizers 0.19.1
389
+ ","{""id"": ""Leeyuyu/idefics3-llama-thyroid"", ""author"": ""Leeyuyu"", ""sha"": ""07c57799e684b04d136b7bef64820ae89696c60a"", ""last_modified"": ""2024-09-09 02:21:46+00:00"", ""created_at"": ""2024-09-09 02:21:42+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""safetensors"", ""generated_from_trainer"", ""base_model:HuggingFaceM4/Idefics3-8B-Llama3"", ""base_model:finetune:HuggingFaceM4/Idefics3-8B-Llama3"", ""license:apache-2.0"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: HuggingFaceM4/Idefics3-8B-Llama3\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: idefics3-llama-thyroid\n results: []"", ""widget_data"": null, ""model_index"": [{""name"": ""idefics3-llama-thyroid"", ""results"": []}], ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='training_args.bin', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-09-09 02:21:46+00:00"", ""cardData"": ""base_model: HuggingFaceM4/Idefics3-8B-Llama3\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: idefics3-llama-thyroid\n results: []"", ""transformersInfo"": null, ""_id"": ""66de5bb6b4d4e7827a332a05"", ""modelId"": ""Leeyuyu/idefics3-llama-thyroid"", ""usedStorage"": 83955400}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=Leeyuyu/idefics3-llama-thyroid&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BLeeyuyu%2Fidefics3-llama-thyroid%5D(%2FLeeyuyu%2Fidefics3-llama-thyroid)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
390
+ Clark12/POS1,"---
391
+ license: mit
392
+ datasets:
393
+ - fka/awesome-chatgpt-prompts
394
+ language:
395
+ - en
396
+ metrics:
397
+ - accuracy
398
+ pipeline_tag: question-answering
399
+ tags:
400
+ - finance
401
+ base_model:
402
+ - HuggingFaceM4/Idefics3-8B-Llama3
403
+ ---
404
+ # Model Card for Model ID
405
+
406
+ <!-- Provide a quick summary of what the model is/does. -->
407
+
408
+ This modelcard aims to be a base template for new models. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/modelcard_template.md?plain=1).
409
+
410
+ ## Model Details
411
+
412
+ ### Model Description
413
+
414
+ <!-- Provide a longer summary of what this model is. -->
415
+
416
+
417
+
418
+ - **Developed by:** [More Information Needed]
419
+ - **Funded by [optional]:** [More Information Needed]
420
+ - **Shared by [optional]:** [More Information Needed]
421
+ - **Model type:** [More Information Needed]
422
+ - **Language(s) (NLP):** [More Information Needed]
423
+ - **License:** [More Information Needed]
424
+ - **Finetuned from model [optional]:** [More Information Needed]
425
+
426
+ ### Model Sources [optional]
427
+
428
+ <!-- Provide the basic links for the model. -->
429
+
430
+ - **Repository:** [More Information Needed]
431
+ - **Paper [optional]:** [More Information Needed]
432
+ - **Demo [optional]:** [More Information Needed]
433
+
434
+ ## Uses
435
+
436
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
437
+
438
+ ### Direct Use
439
+
440
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
441
+
442
+ [More Information Needed]
443
+
444
+ ### Downstream Use [optional]
445
+
446
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
447
+
448
+ [More Information Needed]
449
+
450
+ ### Out-of-Scope Use
451
+
452
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
453
+
454
+ [More Information Needed]
455
+
456
+ ## Bias, Risks, and Limitations
457
+
458
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
459
+
460
+ [More Information Needed]
461
+
462
+ ### Recommendations
463
+
464
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
465
+
466
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
467
+
468
+ ## How to Get Started with the Model
469
+
470
+ Use the code below to get started with the model.
471
+
472
+ [More Information Needed]
473
+
474
+ ## Training Details
475
+
476
+ ### Training Data
477
+
478
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
479
+
480
+ [More Information Needed]
481
+
482
+ ### Training Procedure
483
+
484
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
485
+
486
+ #### Preprocessing [optional]
487
+
488
+ [More Information Needed]
489
+
490
+
491
+ #### Training Hyperparameters
492
+
493
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
494
+
495
+ #### Speeds, Sizes, Times [optional]
496
+
497
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
498
+
499
+ [More Information Needed]
500
+
501
+ ## Evaluation
502
+
503
+ <!-- This section describes the evaluation protocols and provides the results. -->
504
+
505
+ ### Testing Data, Factors & Metrics
506
+
507
+ #### Testing Data
508
+
509
+ <!-- This should link to a Dataset Card if possible. -->
510
+
511
+ [More Information Needed]
512
+
513
+ #### Factors
514
+
515
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
516
+
517
+ [More Information Needed]
518
+
519
+ #### Metrics
520
+
521
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
522
+
523
+ [More Information Needed]
524
+
525
+ ### Results
526
+
527
+ [More Information Needed]
528
+
529
+ #### Summary
530
+
531
+
532
+
533
+ ## Model Examination [optional]
534
+
535
+ <!-- Relevant interpretability work for the model goes here -->
536
+
537
+ [More Information Needed]
538
+
539
+ ## Environmental Impact
540
+
541
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
542
+
543
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
544
+
545
+ - **Hardware Type:** [More Information Needed]
546
+ - **Hours used:** [More Information Needed]
547
+ - **Cloud Provider:** [More Information Needed]
548
+ - **Compute Region:** [More Information Needed]
549
+ - **Carbon Emitted:** [More Information Needed]
550
+
551
+ ## Technical Specifications [optional]
552
+
553
+ ### Model Architecture and Objective
554
+
555
+ [More Information Needed]
556
+
557
+ ### Compute Infrastructure
558
+
559
+ [More Information Needed]
560
+
561
+ #### Hardware
562
+
563
+ [More Information Needed]
564
+
565
+ #### Software
566
+
567
+ [More Information Needed]
568
+
569
+ ## Citation [optional]
570
+
571
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
572
+
573
+ **BibTeX:**
574
+
575
+ [More Information Needed]
576
+
577
+ **APA:**
578
+
579
+ [More Information Needed]
580
+
581
+ ## Glossary [optional]
582
+
583
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
584
+
585
+ [More Information Needed]
586
+
587
+ ## More Information [optional]
588
+
589
+ [More Information Needed]
590
+
591
+ ## Model Card Authors [optional]
592
+
593
+ [More Information Needed]
594
+
595
+ ## Model Card Contact
596
+
597
+ [More Information Needed]","{""id"": ""Clark12/POS1"", ""author"": ""Clark12"", ""sha"": ""a7946677a9b1bfaf17219c181dc137a098427938"", ""last_modified"": ""2024-09-20 20:39:22+00:00"", ""created_at"": ""2024-09-20 20:26:49+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""finance"", ""question-answering"", ""en"", ""dataset:fka/awesome-chatgpt-prompts"", ""arxiv:1910.09700"", ""base_model:HuggingFaceM4/Idefics3-8B-Llama3"", ""base_model:finetune:HuggingFaceM4/Idefics3-8B-Llama3"", ""license:mit"", ""region:us""], ""pipeline_tag"": ""question-answering"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- HuggingFaceM4/Idefics3-8B-Llama3\ndatasets:\n- fka/awesome-chatgpt-prompts\nlanguage:\n- en\nlicense: mit\nmetrics:\n- accuracy\npipeline_tag: question-answering\ntags:\n- finance"", ""widget_data"": [{""text"": ""Where do I live?"", ""context"": ""My name is Wolfgang and I live in Berlin""}, {""text"": ""Where do I live?"", ""context"": ""My name is Sarah and I live in London""}, {""text"": ""What's my name?"", ""context"": ""My name is Clara and I live in Berkeley.""}, {""text"": ""Which name is also used to describe the Amazon rainforest in English?"", ""context"": ""The Amazon rainforest (Portuguese: Floresta Amaz\u00f4nica or Amaz\u00f4nia; Spanish: Selva Amaz\u00f3nica, Amazon\u00eda or usually Amazonia; French: For\u00eat amazonienne; Dutch: Amazoneregenwoud), also known in English as Amazonia or the Amazon Jungle, is a moist broadleaf forest that covers most of the Amazon basin of South America. This basin encompasses 7,000,000 square kilometres (2,700,000 sq mi), of which 5,500,000 square kilometres (2,100,000 sq mi) are covered by the rainforest. This region includes territory belonging to nine nations. The majority of the forest is contained within Brazil, with 60% of the rainforest, followed by Peru with 13%, Colombia with 10%, and with minor amounts in Venezuela, Ecuador, Bolivia, Guyana, Suriname and French Guiana. States or departments in four nations contain \""Amazonas\"" in their names. The Amazon represents over half of the planet's remaining rainforests, and comprises the largest and most biodiverse tract of tropical rainforest in the world, with an estimated 390 billion individual trees divided into 16,000 species.""}], ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-09-20 20:39:22+00:00"", ""cardData"": ""base_model:\n- HuggingFaceM4/Idefics3-8B-Llama3\ndatasets:\n- fka/awesome-chatgpt-prompts\nlanguage:\n- en\nlicense: mit\nmetrics:\n- accuracy\npipeline_tag: question-answering\ntags:\n- finance"", ""transformersInfo"": null, ""_id"": ""66edda89e01b8de82fca1c0a"", ""modelId"": ""Clark12/POS1"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=Clark12/POS1&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BClark12%2FPOS1%5D(%2FClark12%2FPOS1)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
598
+ https://huggingface.co/Minthy/Torii_Gate_v0.1_alpha,N/A,N/A,1,,0,,0,,0,,0,,0
599
+ slezki/assistant_mage,"---
600
+ license: mit
601
+ language:
602
+ - ru
603
+ base_model:
604
+ - HuggingFaceM4/Idefics3-8B-Llama3
605
+ library_name: transformers
606
+ ---","{""id"": ""slezki/assistant_mage"", ""author"": ""slezki"", ""sha"": ""443b97c6ecc62670715384b58c9ecaef3d9cd395"", ""last_modified"": ""2024-09-24 14:54:52+00:00"", ""created_at"": ""2024-09-24 14:50:29+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""ru"", ""base_model:HuggingFaceM4/Idefics3-8B-Llama3"", ""base_model:finetune:HuggingFaceM4/Idefics3-8B-Llama3"", ""license:mit"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- HuggingFaceM4/Idefics3-8B-Llama3\nlanguage:\n- ru\nlibrary_name: transformers\nlicense: mit"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": {""auto_model"": ""AutoModel"", ""custom_class"": null, ""pipeline_tag"": null, ""processor"": null}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-09-24 14:54:52+00:00"", ""cardData"": ""base_model:\n- HuggingFaceM4/Idefics3-8B-Llama3\nlanguage:\n- ru\nlibrary_name: transformers\nlicense: mit"", ""transformersInfo"": {""auto_model"": ""AutoModel"", ""custom_class"": null, ""pipeline_tag"": null, ""processor"": null}, ""_id"": ""66f2d1b55ddc45b044e2c53f"", ""modelId"": ""slezki/assistant_mage"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=slezki/assistant_mage&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bslezki%2Fassistant_mage%5D(%2Fslezki%2Fassistant_mage)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
607
+ https://huggingface.co/Minthy/ToriiGate-v0.2,N/A,N/A,1,,0,,0,,0,,0,,0
InstantID_finetunes_20250425_041137.csv_finetunes_20250425_041137.csv ADDED
@@ -0,0 +1,129 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ InstantX/InstantID,"---
3
+ license: apache-2.0
4
+ language:
5
+ - en
6
+ library_name: diffusers
7
+ pipeline_tag: text-to-image
8
+ ---
9
+
10
+ # InstantID Model Card
11
+
12
+ <div align=""center"">
13
+
14
+ [**Project Page**](https://instantid.github.io/) **|** [**Paper**](https://arxiv.org/abs/2401.07519) **|** [**Code**](https://github.com/InstantID/InstantID) **|** [🤗 **Gradio demo**](https://huggingface.co/spaces/InstantX/InstantID)
15
+
16
+
17
+ </div>
18
+
19
+ ## Introduction
20
+
21
+ InstantID is a new state-of-the-art tuning-free method to achieve ID-Preserving generation with only single image, supporting various downstream tasks.
22
+
23
+ <div align=""center"">
24
+ <img src='examples/applications.png'>
25
+ </div>
26
+
27
+
28
+ ## Usage
29
+
30
+ You can directly download the model in this repository.
31
+ You also can download the model in python script:
32
+
33
+ ```python
34
+ from huggingface_hub import hf_hub_download
35
+ hf_hub_download(repo_id=""InstantX/InstantID"", filename=""ControlNetModel/config.json"", local_dir=""./checkpoints"")
36
+ hf_hub_download(repo_id=""InstantX/InstantID"", filename=""ControlNetModel/diffusion_pytorch_model.safetensors"", local_dir=""./checkpoints"")
37
+ hf_hub_download(repo_id=""InstantX/InstantID"", filename=""ip-adapter.bin"", local_dir=""./checkpoints"")
38
+ ```
39
+
40
+ For face encoder, you need to manutally download via this [URL](https://github.com/deepinsight/insightface/issues/1896#issuecomment-1023867304) to `models/antelopev2`.
41
+
42
+ ```python
43
+ # !pip install opencv-python transformers accelerate insightface
44
+ import diffusers
45
+ from diffusers.utils import load_image
46
+ from diffusers.models import ControlNetModel
47
+
48
+ import cv2
49
+ import torch
50
+ import numpy as np
51
+ from PIL import Image
52
+
53
+ from insightface.app import FaceAnalysis
54
+ from pipeline_stable_diffusion_xl_instantid import StableDiffusionXLInstantIDPipeline, draw_kps
55
+
56
+ # prepare 'antelopev2' under ./models
57
+ app = FaceAnalysis(name='antelopev2', root='./', providers=['CUDAExecutionProvider', 'CPUExecutionProvider'])
58
+ app.prepare(ctx_id=0, det_size=(640, 640))
59
+
60
+ # prepare models under ./checkpoints
61
+ face_adapter = f'./checkpoints/ip-adapter.bin'
62
+ controlnet_path = f'./checkpoints/ControlNetModel'
63
+
64
+ # load IdentityNet
65
+ controlnet = ControlNetModel.from_pretrained(controlnet_path, torch_dtype=torch.float16)
66
+
67
+ pipe = StableDiffusionXLInstantIDPipeline.from_pretrained(
68
+ ... ""stabilityai/stable-diffusion-xl-base-1.0"", controlnet=controlnet, torch_dtype=torch.float16
69
+ ... )
70
+ pipe.cuda()
71
+
72
+ # load adapter
73
+ pipe.load_ip_adapter_instantid(face_adapter)
74
+ ```
75
+
76
+ Then, you can customized your own face images
77
+
78
+ ```python
79
+ # load an image
80
+ image = load_image(""your-example.jpg"")
81
+
82
+ # prepare face emb
83
+ face_info = app.get(cv2.cvtColor(np.array(face_image), cv2.COLOR_RGB2BGR))
84
+ face_info = sorted(face_info, key=lambda x:(x['bbox'][2]-x['bbox'][0])*x['bbox'][3]-x['bbox'][1])[-1] # only use the maximum face
85
+ face_emb = face_info['embedding']
86
+ face_kps = draw_kps(face_image, face_info['kps'])
87
+
88
+ pipe.set_ip_adapter_scale(0.8)
89
+
90
+ prompt = ""analog film photo of a man. faded film, desaturated, 35mm photo, grainy, vignette, vintage, Kodachrome, Lomography, stained, highly detailed, found footage, masterpiece, best quality""
91
+ negative_prompt = ""(lowres, low quality, worst quality:1.2), (text:1.2), watermark, painting, drawing, illustration, glitch, deformed, mutated, cross-eyed, ugly, disfigured (lowres, low quality, worst quality:1.2), (text:1.2), watermark, painting, drawing, illustration, glitch,deformed, mutated, cross-eyed, ugly, disfigured""
92
+
93
+ # generate image
94
+ image = pipe(
95
+ ... prompt, image_embeds=face_emb, image=face_kps, controlnet_conditioning_scale=0.8
96
+ ... ).images[0]
97
+ ```
98
+
99
+ For more details, please follow the instructions in our [GitHub repository](https://github.com/InstantID/InstantID).
100
+
101
+ ## Usage Tips
102
+ 1. If you're not satisfied with the similarity, try to increase the weight of ""IdentityNet Strength"" and ""Adapter Strength"".
103
+ 2. If you feel that the saturation is too high, first decrease the Adapter strength. If it is still too high, then decrease the IdentityNet strength.
104
+ 3. If you find that text control is not as expected, decrease Adapter strength.
105
+ 4. If you find that realistic style is not good enough, go for our Github repo and use a more realistic base model.
106
+
107
+ ## Demos
108
+
109
+ <div align=""center"">
110
+ <img src='examples/0.png'>
111
+ </div>
112
+
113
+ <div align=""center"">
114
+ <img src='examples/1.png'>
115
+ </div>
116
+
117
+ ## Disclaimer
118
+
119
+ This project is released under Apache License and aims to positively impact the field of AI-driven image generation. Users are granted the freedom to create images using this tool, but they are obligated to comply with local laws and utilize it responsibly. The developers will not assume any responsibility for potential misuse by users.
120
+
121
+ ## Citation
122
+ ```bibtex
123
+ @article{wang2024instantid,
124
+ title={InstantID: Zero-shot Identity-Preserving Generation in Seconds},
125
+ author={Wang, Qixun and Bai, Xu and Wang, Haofan and Qin, Zekui and Chen, Anthony},
126
+ journal={arXiv preprint arXiv:2401.07519},
127
+ year={2024}
128
+ }
129
+ ```","{""id"": ""InstantX/InstantID"", ""author"": ""InstantX"", ""sha"": ""57b32dfee076092ad2930c71fd6d439c2c3b1820"", ""last_modified"": ""2024-01-22 09:43:05+00:00"", ""created_at"": ""2024-01-19 11:52:05+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 87716, ""downloads_all_time"": null, ""likes"": 783, ""library_name"": ""diffusers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""diffusers"", ""safetensors"", ""text-to-image"", ""en"", ""arxiv:2401.07519"", ""license:apache-2.0"", ""region:us""], ""pipeline_tag"": ""text-to-image"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""language:\n- en\nlibrary_name: diffusers\nlicense: apache-2.0\npipeline_tag: text-to-image"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ControlNetModel/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ControlNetModel/diffusion_pytorch_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='examples/0.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='examples/1.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='examples/applications.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ip-adapter.bin', size=None, blob_id=None, lfs=None)""], ""spaces"": [""InstantX/InstantID"", ""Nymbo/image_gen_supaqueue"", ""Fucius/OMG-InstantID"", ""ddosxd/InstantID"", ""LPDoctor/InstantID.AIPro"", ""jiaxiangc/res-adapter"", ""cocktailpeanut/InstantID"", ""charlieguo610/InstantID"", ""waloneai/InstantAIPortrait"", ""cocktailpeanut/InstantID2"", ""jcudit/InstantID2"", ""CodeScooper/InstantX-InstantID"", ""batoon/InstantID"", ""allAI-tools/InstantID2"", ""Tototo1394/InstantX-InstantID"", ""Veloptesauarzor/InstantX-InstantID"", ""JCTN/InstantID"", ""tsi-org/InstantID"", ""listkun/InstantX-InstantID"", ""Jsachman/InstantX-InstantID"", ""uelordi/InstantID"", ""xiaoshuai11111/InstantX-InstantID"", ""joey1895/InstantX-InstantID"", ""iamshiss/InstantX-test-for-shishishi"", ""canismadjor/InstantX-InstantID"", ""darshcoss/InstantID"", ""nianevermore/InstantX-InstantID"", ""seawolf2357/vidiid"", ""vivek6900/InstantX-InstantID"", ""Cronix90/InstantX-InstantID"", ""dreamlord1995/InstantX-InstantID"", ""Irishcoder/InstantX-InstantID"", ""cement-dev/InstantX-InstantID"", ""facehugger222/h"", ""Apsoedarsono/InstantX-InstantID"", ""Tpie333/InstantX-InstantID"", ""GardenXas/InstantX-InstantID"", ""Crack506/InstantX-InstantID"", ""thekubist/InstantX-InstantID"", ""ligan/InstantX-InstantID"", ""puseletso55/InstantX-InstantID"", ""xiaoli12345/InstantX-InstantID"", ""yuxh1996/InstantID.AIPro"", ""0xZWang/InstantX-InstantID"", ""Seanwinners/InstantX-InstantID"", ""TopStreetFights/InstantX-InstantID"", ""lucky95271/InstantX-InstantID"", ""Dineth1222/_generete_image_with_nova"", ""Dineth1222/nova_image_gen"", ""Dineth1222/imageg_generete_with_nova"", ""jensinjames/InstantX-InstantID"", ""hatkarsaheb1776/InstantX-InstantID"", ""chaim/InstantX-InstantID"", ""h20ahmadi/InstantX-InstantID"", ""hhhhhhhhdss/InstantX-InstantID"", ""anandx/InstantX-InstantID"", ""opq741/InstantX-InstantID"", ""Dynexcorp/InstantX-InstantID"", ""fujue/InstantX-InstantID"", ""ajcdp/InstantX-InstantID"", ""canerdogan/InstantX-InstantID"", ""jw1900/InstantID"", ""asqwerty/InstantX-InstantID"", ""dd890/InstantX-InstantID"", ""lightsnail/InstantX-InstantID"", ""Mago-pio/InstantX-InstantID"", ""Kthkng/InstantX-InstantID"", ""JLewisT/InstantX-InstantID"", ""Erwnna/InstantX-InstantID"", ""Shanedignan/InstantX-InstantID"", ""Nickegan/InstantX-InstantID"", ""r2hu1/InstantX-InstantID"", ""jaisonkerala1/InstantX-InstantID"", ""Spongenuity/SomFingImade"", ""Biguenda/InstantX-InstantID"", ""Spongenuity/iMadeAFing"", ""Rumman157/InstantX-InstantID"", ""brianying/InstantID"", ""IsaacRDGZ05/InstantX-InstantID"", ""ThaDonald/InstantID69"", ""osmunphotography/InstantX-InstantID5"", ""CJAlos/InstantID2"", ""Dreamacus/InstantX-InstantID"", ""3bodyproblem/InstantX-InstantID"", ""Samarth0710/InstantX-InstantID"", ""Bigdaddyborch/InstantX-InstantID"", ""findpavan/InstantX-InstantID"", ""rahulbomnalli/InstantX-InstantID"", ""pranay143342/InstantX-InstantID"", ""HSxxx/InstantX-InstantID"", ""helloraj/InstantX-InstantID"", ""Tonyaispuro/InstantX-InstantID"", ""Gumm1/InstantX-InstantID"", ""wokebo/InstantX-InstantID"", ""clarkasian/InstantX-InstantID"", ""Nicoriba/InstantX-InstantID"", ""letskillgod/InstantX-InstantID"", ""ZestySalsa/InstantX-InstantID"", ""natti0170/InstantX-InstantID"", ""namuit/InstantID""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-01-22 09:43:05+00:00"", ""cardData"": ""language:\n- en\nlibrary_name: diffusers\nlicense: apache-2.0\npipeline_tag: text-to-image"", ""transformersInfo"": null, ""_id"": ""65aa62652f560c70ffe691a9"", ""modelId"": ""InstantX/InstantID"", ""usedStorage"": 4263855348}",0,,0,"https://huggingface.co/loliOppai/KyOresu, https://huggingface.co/zz001/llll, https://huggingface.co/zz001/45435, https://huggingface.co/lylosn/plum, https://huggingface.co/coversia21/GermanGarmendia, https://huggingface.co/LegoClipStars/GetBlakeBlakeMyers",6,,0,,0,"CodeScooper/InstantX-InstantID, Fucius/OMG-InstantID, InstantX/InstantID, LPDoctor/InstantID.AIPro, Nymbo/image_gen_supaqueue, charlieguo610/InstantID, cocktailpeanut/InstantID, cocktailpeanut/InstantID2, ddosxd/InstantID, huggingface/InferenceSupport/discussions/new?title=InstantX/InstantID&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BInstantX%2FInstantID%5D(%2FInstantX%2FInstantID)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, jiaxiangc/res-adapter, joey1895/InstantX-InstantID, listkun/InstantX-InstantID",13
InternVL2-Llama3-76B_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv ADDED
@@ -0,0 +1,652 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ OpenGVLab/InternVL2-Llama3-76B,"---
3
+ license: llama3
4
+ pipeline_tag: image-text-to-text
5
+ library_name: transformers
6
+ base_model:
7
+ - OpenGVLab/InternViT-6B-448px-V1-5
8
+ - NousResearch/Hermes-2-Theta-Llama-3-70B
9
+ new_version: OpenGVLab/InternVL2_5-78B
10
+ base_model_relation: merge
11
+ language:
12
+ - multilingual
13
+ tags:
14
+ - internvl
15
+ - custom_code
16
+ ---
17
+
18
+ # InternVL2-Llama3-76B
19
+
20
+ [\[📂 GitHub\]](https://github.com/OpenGVLab/InternVL) [\[📜 InternVL 1.0\]](https://huggingface.co/papers/2312.14238) [\[📜 InternVL 1.5\]](https://huggingface.co/papers/2404.16821) [\[📜 Mini-InternVL\]](https://arxiv.org/abs/2410.16261) [\[📜 InternVL 2.5\]](https://huggingface.co/papers/2412.05271)
21
+
22
+ [\[🆕 Blog\]](https://internvl.github.io/blog/) [\[🗨️ Chat Demo\]](https://internvl.opengvlab.com/) [\[🤗 HF Demo\]](https://huggingface.co/spaces/OpenGVLab/InternVL) [\[🚀 Quick Start\]](#quick-start) [\[📖 Documents\]](https://internvl.readthedocs.io/en/latest/)
23
+
24
+ <div align=""center"">
25
+ <img width=""500"" alt=""image"" src=""https://cdn-uploads.huggingface.co/production/uploads/64006c09330a45b03605bba3/zJsd2hqd3EevgXo6fNgC-.png"">
26
+ </div>
27
+
28
+ ## Introduction
29
+
30
+ We are excited to announce the release of InternVL 2.0, the latest addition to the InternVL series of multimodal large language models. InternVL 2.0 features a variety of **instruction-tuned models**, ranging from 1 billion to 108 billion parameters. This repository contains the instruction-tuned InternVL2-Llama3-76B model.
31
+
32
+ Compared to the state-of-the-art open-source multimodal large language models, InternVL 2.0 surpasses most open-source models. It demonstrates competitive performance on par with proprietary commercial models across various capabilities, including document and chart comprehension, infographics QA, scene text understanding and OCR tasks, scientific and mathematical problem solving, as well as cultural understanding and integrated multimodal capabilities.
33
+
34
+ InternVL 2.0 is trained with an 8k context window and utilizes training data consisting of long texts, multiple images, and videos, significantly improving its ability to handle these types of inputs compared to InternVL 1.5. For more details, please refer to our [blog](https://internvl.github.io/blog/2024-07-02-InternVL-2.0/) and [GitHub](https://github.com/OpenGVLab/InternVL).
35
+
36
+ | Model Name | Vision Part | Language Part | HF Link | MS Link |
37
+ | :------------------: | :---------------------------------------------------------------------------------: | :------------------------------------------------------------------------------------------: | :--------------------------------------------------------------: | :--------------------------------------------------------------------: |
38
+ | InternVL2-1B | [InternViT-300M-448px](https://huggingface.co/OpenGVLab/InternViT-300M-448px) | [Qwen2-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2-0.5B-Instruct) | [🤗 link](https://huggingface.co/OpenGVLab/InternVL2-1B) | [🤖 link](https://modelscope.cn/models/OpenGVLab/InternVL2-1B) |
39
+ | InternVL2-2B | [InternViT-300M-448px](https://huggingface.co/OpenGVLab/InternViT-300M-448px) | [internlm2-chat-1_8b](https://huggingface.co/internlm/internlm2-chat-1_8b) | [🤗 link](https://huggingface.co/OpenGVLab/InternVL2-2B) | [🤖 link](https://modelscope.cn/models/OpenGVLab/InternVL2-2B) |
40
+ | InternVL2-4B | [InternViT-300M-448px](https://huggingface.co/OpenGVLab/InternViT-300M-448px) | [Phi-3-mini-128k-instruct](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct) | [🤗 link](https://huggingface.co/OpenGVLab/InternVL2-4B) | [🤖 link](https://modelscope.cn/models/OpenGVLab/InternVL2-4B) |
41
+ | InternVL2-8B | [InternViT-300M-448px](https://huggingface.co/OpenGVLab/InternViT-300M-448px) | [internlm2_5-7b-chat](https://huggingface.co/internlm/internlm2_5-7b-chat) | [🤗 link](https://huggingface.co/OpenGVLab/InternVL2-8B) | [🤖 link](https://modelscope.cn/models/OpenGVLab/InternVL2-8B) |
42
+ | InternVL2-26B | [InternViT-6B-448px-V1-5](https://huggingface.co/OpenGVLab/InternViT-6B-448px-V1-5) | [internlm2-chat-20b](https://huggingface.co/internlm/internlm2-chat-20b) | [🤗 link](https://huggingface.co/OpenGVLab/InternVL2-26B) | [🤖 link](https://modelscope.cn/models/OpenGVLab/InternVL2-26B) |
43
+ | InternVL2-40B | [InternViT-6B-448px-V1-5](https://huggingface.co/OpenGVLab/InternViT-6B-448px-V1-5) | [Nous-Hermes-2-Yi-34B](https://huggingface.co/NousResearch/Nous-Hermes-2-Yi-34B) | [🤗 link](https://huggingface.co/OpenGVLab/InternVL2-40B) | [🤖 link](https://modelscope.cn/models/OpenGVLab/InternVL2-40B) |
44
+ | InternVL2-Llama3-76B | [InternViT-6B-448px-V1-5](https://huggingface.co/OpenGVLab/InternViT-6B-448px-V1-5) | [Hermes-2-Theta-Llama-3-70B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-70B) | [🤗 link](https://huggingface.co/OpenGVLab/InternVL2-Llama3-76B) | [🤖 link](https://modelscope.cn/models/OpenGVLab/InternVL2-Llama3-76B) |
45
+
46
+ ## Model Details
47
+
48
+ InternVL 2.0 is a multimodal large language model series, featuring models of various sizes. For each size, we release instruction-tuned models optimized for multimodal tasks. InternVL2-Llama3-76B consists of [InternViT-6B-448px-V1-5](https://huggingface.co/OpenGVLab/InternViT-6B-448px-V1-5), an MLP projector, and [Hermes-2-Theta-Llama-3-70B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-70B).
49
+
50
+ ## Performance
51
+
52
+ ### Image Benchmarks
53
+
54
+ | Benchmark | GPT-4o-20240513 | Claude3.5-Sonnet | InternVL2-40B | InternVL2-Llama3-76B |
55
+ | :--------------------------: | :-------------: | :--------------: | :-----------: | :------------------: |
56
+ | Model Size | - | - | 40B | 76B |
57
+ | | | | | |
58
+ | DocVQA<sub>test</sub> | 92.8 | 95.2 | 93.9 | 94.1 |
59
+ | ChartQA<sub>test</sub> | 85.7 | 90.8 | 86.2 | 88.4 |
60
+ | InfoVQA<sub>test</sub> | - | - | 78.7 | 82.0 |
61
+ | TextVQA<sub>val</sub> | - | - | 83.0 | 84.4 |
62
+ | OCRBench | 736 | 788 | 837 | 839 |
63
+ | MME<sub>sum</sub> | 2328.7 | 1920.0 | 2315.0 | 2414.7 |
64
+ | RealWorldQA | 75.4 | 60.1 | 71.8 | 72.2 |
65
+ | AI2D<sub>test</sub> | 94.2 | 94.7 | 87.1 | 87.6 |
66
+ | MMMU<sub>val</sub> | 69.1 | 68.3 | 55.2 | 58.2 |
67
+ | MMBench-EN<sub>test</sub> | 83.4 | 79.7 | 86.8 | 86.5 |
68
+ | MMBench-CN<sub>test</sub> | 82.1 | 80.7 | 86.5 | 86.3 |
69
+ | CCBench<sub>dev</sub> | 71.2 | 54.1 | 80.6 | 81.0 |
70
+ | MMVet<sub>GPT-4-0613</sub> | - | - | 68.5 | 69.8 |
71
+ | MMVet<sub>GPT-4-Turbo</sub> | 69.1 | 66.0 | 65.5 | 65.7 |
72
+ | SEED-Image | 77.1 | - | 78.2 | 78.2 |
73
+ | HallBench<sub>avg</sub> | 55.0 | 49.9 | 56.9 | 55.2 |
74
+ | MathVista<sub>testmini</sub> | 63.8 | 67.7 | 63.7 | 65.5 |
75
+ | OpenCompass<sub>avg</sub> | 69.9 | 67.9 | 69.7 | 71.0 |
76
+
77
+ - For more details and evaluation reproduction, please refer to our [Evaluation Guide](https://internvl.readthedocs.io/en/latest/internvl2.0/evaluation.html).
78
+
79
+ - We simultaneously use [InternVL](https://github.com/OpenGVLab/InternVL) and [VLMEvalKit](https://github.com/open-compass/VLMEvalKit) repositories for model evaluation. Specifically, the results reported for DocVQA, ChartQA, InfoVQA, TextVQA, MME, AI2D, MMBench, CCBench, MMVet (GPT-4-0613), and SEED-Image were tested using the InternVL repository. MMMU, OCRBench, RealWorldQA, HallBench, MMVet (GPT-4-Turbo), and MathVista were evaluated using the VLMEvalKit.
80
+
81
+ ### Video Benchmarks
82
+
83
+ | Benchmark | GPT-4o | GPT-4V | Gemini-Pro-1.5 | InternVL2-40B | InternVL2-Llama3-76B |
84
+ | :-------------------------: | :----: | :----: | :------------: | :-----------: | :------------------: |
85
+ | Model Size | - | - | - | 40B | 76B |
86
+ | | | | | | |
87
+ | MVBench | - | - | - | 72.5 | 69.6 |
88
+ | MMBench-Video<sub>8f</sub> | 1.62 | 1.53 | 1.30 | 1.32 | 1.37 |
89
+ | MMBench-Video<sub>16f</sub> | 1.86 | 1.68 | 1.60 | 1.45 | 1.52 |
90
+ | Video-MME<br>w/o subs | 71.9 | 59.9 | 75.0 | 61.2 | 61.2 |
91
+ | Video-MME<br>w subs | 77.2 | 63.3 | 81.3 | 62.4 | 62.8 |
92
+
93
+ - We evaluate our models on MVBench and Video-MME by extracting 16 frames from each video, and each frame was resized to a 448x448 image.
94
+
95
+ ### Grounding Benchmarks
96
+
97
+ | Model | avg. | RefCOCO<br>(val) | RefCOCO<br>(testA) | RefCOCO<br>(testB) | RefCOCO+<br>(val) | RefCOCO+<br>(testA) | RefCOCO+<br>(testB) | RefCOCO‑g<br>(val) | RefCOCO‑g<br>(test) |
98
+ | :----------------------------: | :--: | :--------------: | :----------------: | :----------------: | :---------------: | :-----------------: | :-----------------: | :----------------: | :-----------------: |
99
+ | UNINEXT-H<br>(Specialist SOTA) | 88.9 | 92.6 | 94.3 | 91.5 | 85.2 | 89.6 | 79.8 | 88.7 | 89.4 |
100
+ | | | | | | | | | | |
101
+ | Mini-InternVL-<br>Chat-2B-V1-5 | 75.8 | 80.7 | 86.7 | 72.9 | 72.5 | 82.3 | 60.8 | 75.6 | 74.9 |
102
+ | Mini-InternVL-<br>Chat-4B-V1-5 | 84.4 | 88.0 | 91.4 | 83.5 | 81.5 | 87.4 | 73.8 | 84.7 | 84.6 |
103
+ | InternVL‑Chat‑V1‑5 | 88.8 | 91.4 | 93.7 | 87.1 | 87.0 | 92.3 | 80.9 | 88.5 | 89.3 |
104
+ | | | | | | | | | | |
105
+ | InternVL2‑1B | 79.9 | 83.6 | 88.7 | 79.8 | 76.0 | 83.6 | 67.7 | 80.2 | 79.9 |
106
+ | InternVL2‑2B | 77.7 | 82.3 | 88.2 | 75.9 | 73.5 | 82.8 | 63.3 | 77.6 | 78.3 |
107
+ | InternVL2‑4B | 84.4 | 88.5 | 91.2 | 83.9 | 81.2 | 87.2 | 73.8 | 84.6 | 84.6 |
108
+ | InternVL2‑8B | 82.9 | 87.1 | 91.1 | 80.7 | 79.8 | 87.9 | 71.4 | 82.7 | 82.7 |
109
+ | InternVL2‑26B | 88.5 | 91.2 | 93.3 | 87.4 | 86.8 | 91.0 | 81.2 | 88.5 | 88.6 |
110
+ | InternVL2‑40B | 90.3 | 93.0 | 94.7 | 89.2 | 88.5 | 92.8 | 83.6 | 90.3 | 90.6 |
111
+ | InternVL2-<br>Llama3‑76B | 90.0 | 92.2 | 94.8 | 88.4 | 88.8 | 93.1 | 82.8 | 89.5 | 90.3 |
112
+
113
+ - We use the following prompt to evaluate InternVL's grounding ability: `Please provide the bounding box coordinates of the region this sentence describes: <ref>{}</ref>`
114
+
115
+ Limitations: Although we have made efforts to ensure the safety of the model during the training process and to encourage the model to generate text that complies with ethical and legal requirements, the model may still produce unexpected outputs due to its size and probabilistic generation paradigm. For example, the generated responses may contain biases, discrimination, or other harmful content. Please do not propagate such content. We are not responsible for any consequences resulting from the dissemination of harmful information.
116
+
117
+ ## Quick Start
118
+
119
+ We provide an example code to run `InternVL2-Llama3-76B` using `transformers`.
120
+
121
+ > Please use transformers>=4.37.2 to ensure the model works normally.
122
+
123
+ ### Model Loading
124
+
125
+ #### 16-bit (bf16 / fp16)
126
+
127
+ ```python
128
+ import torch
129
+ from transformers import AutoTokenizer, AutoModel
130
+ path = ""OpenGVLab/InternVL2-Llama3-76B""
131
+ model = AutoModel.from_pretrained(
132
+ path,
133
+ torch_dtype=torch.bfloat16,
134
+ low_cpu_mem_usage=True,
135
+ use_flash_attn=True,
136
+ trust_remote_code=True).eval().cuda()
137
+ ```
138
+
139
+ #### BNB 8-bit Quantization
140
+
141
+ ```python
142
+ import torch
143
+ from transformers import AutoTokenizer, AutoModel
144
+ path = ""OpenGVLab/InternVL2-Llama3-76B""
145
+ model = AutoModel.from_pretrained(
146
+ path,
147
+ torch_dtype=torch.bfloat16,
148
+ load_in_8bit=True,
149
+ low_cpu_mem_usage=True,
150
+ use_flash_attn=True,
151
+ trust_remote_code=True).eval()
152
+ ```
153
+
154
+ #### Multiple GPUs
155
+
156
+ The reason for writing the code this way is to avoid errors that occur during multi-GPU inference due to tensors not being on the same device. By ensuring that the first and last layers of the large language model (LLM) are on the same device, we prevent such errors.
157
+
158
+ ```python
159
+ import math
160
+ import torch
161
+ from transformers import AutoTokenizer, AutoModel
162
+
163
+ def split_model(model_name):
164
+ device_map = {}
165
+ world_size = torch.cuda.device_count()
166
+ num_layers = {
167
+ 'InternVL2-1B': 24, 'InternVL2-2B': 24, 'InternVL2-4B': 32, 'InternVL2-8B': 32,
168
+ 'InternVL2-26B': 48, 'InternVL2-40B': 60, 'InternVL2-Llama3-76B': 80}[model_name]
169
+ # Since the first GPU will be used for ViT, treat it as half a GPU.
170
+ num_layers_per_gpu = math.ceil(num_layers / (world_size - 0.5))
171
+ num_layers_per_gpu = [num_layers_per_gpu] * world_size
172
+ num_layers_per_gpu[0] = math.ceil(num_layers_per_gpu[0] * 0.5)
173
+ layer_cnt = 0
174
+ for i, num_layer in enumerate(num_layers_per_gpu):
175
+ for j in range(num_layer):
176
+ device_map[f'language_model.model.layers.{layer_cnt}'] = i
177
+ layer_cnt += 1
178
+ device_map['vision_model'] = 0
179
+ device_map['mlp1'] = 0
180
+ device_map['language_model.model.tok_embeddings'] = 0
181
+ device_map['language_model.model.embed_tokens'] = 0
182
+ device_map['language_model.output'] = 0
183
+ device_map['language_model.model.norm'] = 0
184
+ device_map['language_model.model.rotary_emb'] = 0
185
+ device_map['language_model.lm_head'] = 0
186
+ device_map[f'language_model.model.layers.{num_layers - 1}'] = 0
187
+
188
+ return device_map
189
+
190
+ path = ""OpenGVLab/InternVL2-Llama3-76B""
191
+ device_map = split_model('InternVL2-Llama3-76B')
192
+ model = AutoModel.from_pretrained(
193
+ path,
194
+ torch_dtype=torch.bfloat16,
195
+ low_cpu_mem_usage=True,
196
+ use_flash_attn=True,
197
+ trust_remote_code=True,
198
+ device_map=device_map).eval()
199
+ ```
200
+
201
+ ### Inference with Transformers
202
+
203
+ ```python
204
+ import math
205
+ import numpy as np
206
+ import torch
207
+ import torchvision.transforms as T
208
+ from decord import VideoReader, cpu
209
+ from PIL import Image
210
+ from torchvision.transforms.functional import InterpolationMode
211
+ from transformers import AutoModel, AutoTokenizer
212
+
213
+ IMAGENET_MEAN = (0.485, 0.456, 0.406)
214
+ IMAGENET_STD = (0.229, 0.224, 0.225)
215
+
216
+ def build_transform(input_size):
217
+ MEAN, STD = IMAGENET_MEAN, IMAGENET_STD
218
+ transform = T.Compose([
219
+ T.Lambda(lambda img: img.convert('RGB') if img.mode != 'RGB' else img),
220
+ T.Resize((input_size, input_size), interpolation=InterpolationMode.BICUBIC),
221
+ T.ToTensor(),
222
+ T.Normalize(mean=MEAN, std=STD)
223
+ ])
224
+ return transform
225
+
226
+ def find_closest_aspect_ratio(aspect_ratio, target_ratios, width, height, image_size):
227
+ best_ratio_diff = float('inf')
228
+ best_ratio = (1, 1)
229
+ area = width * height
230
+ for ratio in target_ratios:
231
+ target_aspect_ratio = ratio[0] / ratio[1]
232
+ ratio_diff = abs(aspect_ratio - target_aspect_ratio)
233
+ if ratio_diff < best_ratio_diff:
234
+ best_ratio_diff = ratio_diff
235
+ best_ratio = ratio
236
+ elif ratio_diff == best_ratio_diff:
237
+ if area > 0.5 * image_size * image_size * ratio[0] * ratio[1]:
238
+ best_ratio = ratio
239
+ return best_ratio
240
+
241
+ def dynamic_preprocess(image, min_num=1, max_num=12, image_size=448, use_thumbnail=False):
242
+ orig_width, orig_height = image.size
243
+ aspect_ratio = orig_width / orig_height
244
+
245
+ # calculate the existing image aspect ratio
246
+ target_ratios = set(
247
+ (i, j) for n in range(min_num, max_num + 1) for i in range(1, n + 1) for j in range(1, n + 1) if
248
+ i * j <= max_num and i * j >= min_num)
249
+ target_ratios = sorted(target_ratios, key=lambda x: x[0] * x[1])
250
+
251
+ # find the closest aspect ratio to the target
252
+ target_aspect_ratio = find_closest_aspect_ratio(
253
+ aspect_ratio, target_ratios, orig_width, orig_height, image_size)
254
+
255
+ # calculate the target width and height
256
+ target_width = image_size * target_aspect_ratio[0]
257
+ target_height = image_size * target_aspect_ratio[1]
258
+ blocks = target_aspect_ratio[0] * target_aspect_ratio[1]
259
+
260
+ # resize the image
261
+ resized_img = image.resize((target_width, target_height))
262
+ processed_images = []
263
+ for i in range(blocks):
264
+ box = (
265
+ (i % (target_width // image_size)) * image_size,
266
+ (i // (target_width // image_size)) * image_size,
267
+ ((i % (target_width // image_size)) + 1) * image_size,
268
+ ((i // (target_width // image_size)) + 1) * image_size
269
+ )
270
+ # split the image
271
+ split_img = resized_img.crop(box)
272
+ processed_images.append(split_img)
273
+ assert len(processed_images) == blocks
274
+ if use_thumbnail and len(processed_images) != 1:
275
+ thumbnail_img = image.resize((image_size, image_size))
276
+ processed_images.append(thumbnail_img)
277
+ return processed_images
278
+
279
+ def load_image(image_file, input_size=448, max_num=12):
280
+ image = Image.open(image_file).convert('RGB')
281
+ transform = build_transform(input_size=input_size)
282
+ images = dynamic_preprocess(image, image_size=input_size, use_thumbnail=True, max_num=max_num)
283
+ pixel_values = [transform(image) for image in images]
284
+ pixel_values = torch.stack(pixel_values)
285
+ return pixel_values
286
+
287
+ def split_model(model_name):
288
+ device_map = {}
289
+ world_size = torch.cuda.device_count()
290
+ num_layers = {
291
+ 'InternVL2-1B': 24, 'InternVL2-2B': 24, 'InternVL2-4B': 32, 'InternVL2-8B': 32,
292
+ 'InternVL2-26B': 48, 'InternVL2-40B': 60, 'InternVL2-Llama3-76B': 80}[model_name]
293
+ # Since the first GPU will be used for ViT, treat it as half a GPU.
294
+ num_layers_per_gpu = math.ceil(num_layers / (world_size - 0.5))
295
+ num_layers_per_gpu = [num_layers_per_gpu] * world_size
296
+ num_layers_per_gpu[0] = math.ceil(num_layers_per_gpu[0] * 0.5)
297
+ layer_cnt = 0
298
+ for i, num_layer in enumerate(num_layers_per_gpu):
299
+ for j in range(num_layer):
300
+ device_map[f'language_model.model.layers.{layer_cnt}'] = i
301
+ layer_cnt += 1
302
+ device_map['vision_model'] = 0
303
+ device_map['mlp1'] = 0
304
+ device_map['language_model.model.tok_embeddings'] = 0
305
+ device_map['language_model.model.embed_tokens'] = 0
306
+ device_map['language_model.output'] = 0
307
+ device_map['language_model.model.norm'] = 0
308
+ device_map['language_model.model.rotary_emb'] = 0
309
+ device_map['language_model.lm_head'] = 0
310
+ device_map[f'language_model.model.layers.{num_layers - 1}'] = 0
311
+
312
+ return device_map
313
+
314
+ # If you set `load_in_8bit=True`, you will need two 80GB GPUs.
315
+ # If you set `load_in_8bit=False`, you will need at least three 80GB GPUs.
316
+ path = 'OpenGVLab/InternVL2-Llama3-76B'
317
+ device_map = split_model('InternVL2-Llama3-76B')
318
+ model = AutoModel.from_pretrained(
319
+ path,
320
+ torch_dtype=torch.bfloat16,
321
+ load_in_8bit=True,
322
+ low_cpu_mem_usage=True,
323
+ use_flash_attn=True,
324
+ trust_remote_code=True,
325
+ device_map=device_map).eval()
326
+ tokenizer = AutoTokenizer.from_pretrained(path, trust_remote_code=True, use_fast=False)
327
+
328
+ # set the max number of tiles in `max_num`
329
+ pixel_values = load_image('./examples/image1.jpg', max_num=12).to(torch.bfloat16).cuda()
330
+ generation_config = dict(max_new_tokens=1024, do_sample=True)
331
+
332
+ # pure-text conversation (纯文本对话)
333
+ question = 'Hello, who are you?'
334
+ response, history = model.chat(tokenizer, None, question, generation_config, history=None, return_history=True)
335
+ print(f'User: {question}\nAssistant: {response}')
336
+
337
+ question = 'Can you tell me a story?'
338
+ response, history = model.chat(tokenizer, None, question, generation_config, history=history, return_history=True)
339
+ print(f'User: {question}\nAssistant: {response}')
340
+
341
+ # single-image single-round conversation (单图单轮对话)
342
+ question = '<image>\nPlease describe the image shortly.'
343
+ response = model.chat(tokenizer, pixel_values, question, generation_config)
344
+ print(f'User: {question}\nAssistant: {response}')
345
+
346
+ # single-image multi-round conversation (单图多轮对话)
347
+ question = '<image>\nPlease describe the image in detail.'
348
+ response, history = model.chat(tokenizer, pixel_values, question, generation_config, history=None, return_history=True)
349
+ print(f'User: {question}\nAssistant: {response}')
350
+
351
+ question = 'Please write a poem according to the image.'
352
+ response, history = model.chat(tokenizer, pixel_values, question, generation_config, history=history, return_history=True)
353
+ print(f'User: {question}\nAssistant: {response}')
354
+
355
+ # multi-image multi-round conversation, combined images (多图多轮对话，拼接图像)
356
+ pixel_values1 = load_image('./examples/image1.jpg', max_num=12).to(torch.bfloat16).cuda()
357
+ pixel_values2 = load_image('./examples/image2.jpg', max_num=12).to(torch.bfloat16).cuda()
358
+ pixel_values = torch.cat((pixel_values1, pixel_values2), dim=0)
359
+
360
+ question = '<image>\nDescribe the two images in detail.'
361
+ response, history = model.chat(tokenizer, pixel_values, question, generation_config,
362
+ history=None, return_history=True)
363
+ print(f'User: {question}\nAssistant: {response}')
364
+
365
+ question = 'What are the similarities and differences between these two images.'
366
+ response, history = model.chat(tokenizer, pixel_values, question, generation_config,
367
+ history=history, return_history=True)
368
+ print(f'User: {question}\nAssistant: {response}')
369
+
370
+ # multi-image multi-round conversation, separate images (多图多轮对话，独立图像)
371
+ pixel_values1 = load_image('./examples/image1.jpg', max_num=12).to(torch.bfloat16).cuda()
372
+ pixel_values2 = load_image('./examples/image2.jpg', max_num=12).to(torch.bfloat16).cuda()
373
+ pixel_values = torch.cat((pixel_values1, pixel_values2), dim=0)
374
+ num_patches_list = [pixel_values1.size(0), pixel_values2.size(0)]
375
+
376
+ question = 'Image-1: <image>\nImage-2: <image>\nDescribe the two images in detail.'
377
+ response, history = model.chat(tokenizer, pixel_values, question, generation_config,
378
+ num_patches_list=num_patches_list,
379
+ history=None, return_history=True)
380
+ print(f'User: {question}\nAssistant: {response}')
381
+
382
+ question = 'What are the similarities and differences between these two images.'
383
+ response, history = model.chat(tokenizer, pixel_values, question, generation_config,
384
+ num_patches_list=num_patches_list,
385
+ history=history, return_history=True)
386
+ print(f'User: {question}\nAssistant: {response}')
387
+
388
+ # batch inference, single image per sample (单图批处理)
389
+ pixel_values1 = load_image('./examples/image1.jpg', max_num=12).to(torch.bfloat16).cuda()
390
+ pixel_values2 = load_image('./examples/image2.jpg', max_num=12).to(torch.bfloat16).cuda()
391
+ num_patches_list = [pixel_values1.size(0), pixel_values2.size(0)]
392
+ pixel_values = torch.cat((pixel_values1, pixel_values2), dim=0)
393
+
394
+ questions = ['<image>\nDescribe the image in detail.'] * len(num_patches_list)
395
+ responses = model.batch_chat(tokenizer, pixel_values,
396
+ num_patches_list=num_patches_list,
397
+ questions=questions,
398
+ generation_config=generation_config)
399
+ for question, response in zip(questions, responses):
400
+ print(f'User: {question}\nAssistant: {response}')
401
+
402
+ # video multi-round conversation (视频多轮对话)
403
+ def get_index(bound, fps, max_frame, first_idx=0, num_segments=32):
404
+ if bound:
405
+ start, end = bound[0], bound[1]
406
+ else:
407
+ start, end = -100000, 100000
408
+ start_idx = max(first_idx, round(start * fps))
409
+ end_idx = min(round(end * fps), max_frame)
410
+ seg_size = float(end_idx - start_idx) / num_segments
411
+ frame_indices = np.array([
412
+ int(start_idx + (seg_size / 2) + np.round(seg_size * idx))
413
+ for idx in range(num_segments)
414
+ ])
415
+ return frame_indices
416
+
417
+ def load_video(video_path, bound=None, input_size=448, max_num=1, num_segments=32):
418
+ vr = VideoReader(video_path, ctx=cpu(0), num_threads=1)
419
+ max_frame = len(vr) - 1
420
+ fps = float(vr.get_avg_fps())
421
+
422
+ pixel_values_list, num_patches_list = [], []
423
+ transform = build_transform(input_size=input_size)
424
+ frame_indices = get_index(bound, fps, max_frame, first_idx=0, num_segments=num_segments)
425
+ for frame_index in frame_indices:
426
+ img = Image.fromarray(vr[frame_index].asnumpy()).convert('RGB')
427
+ img = dynamic_preprocess(img, image_size=input_size, use_thumbnail=True, max_num=max_num)
428
+ pixel_values = [transform(tile) for tile in img]
429
+ pixel_values = torch.stack(pixel_values)
430
+ num_patches_list.append(pixel_values.shape[0])
431
+ pixel_values_list.append(pixel_values)
432
+ pixel_values = torch.cat(pixel_values_list)
433
+ return pixel_values, num_patches_list
434
+
435
+ video_path = './examples/red-panda.mp4'
436
+ pixel_values, num_patches_list = load_video(video_path, num_segments=8, max_num=1)
437
+ pixel_values = pixel_values.to(torch.bfloat16).cuda()
438
+ video_prefix = ''.join([f'Frame{i+1}: <image>\n' for i in range(len(num_patches_list))])
439
+ question = video_prefix + 'What is the red panda doing?'
440
+ # Frame1: <image>\nFrame2: <image>\n...\nFrame8: <image>\n{question}
441
+ response, history = model.chat(tokenizer, pixel_values, question, generation_config,
442
+ num_patches_list=num_patches_list, history=None, return_history=True)
443
+ print(f'User: {question}\nAssistant: {response}')
444
+
445
+ question = 'Describe this video in detail.'
446
+ response, history = model.chat(tokenizer, pixel_values, question, generation_config,
447
+ num_patches_list=num_patches_list, history=history, return_history=True)
448
+ print(f'User: {question}\nAssistant: {response}')
449
+ ```
450
+
451
+ #### Streaming Output
452
+
453
+ Besides this method, you can also use the following code to get streamed output.
454
+
455
+ ```python
456
+ from transformers import TextIteratorStreamer
457
+ from threading import Thread
458
+
459
+ # Initialize the streamer
460
+ streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True, timeout=10)
461
+ # Define the generation configuration
462
+ generation_config = dict(max_new_tokens=1024, do_sample=False, streamer=streamer)
463
+ # Start the model chat in a separate thread
464
+ thread = Thread(target=model.chat, kwargs=dict(
465
+ tokenizer=tokenizer, pixel_values=pixel_values, question=question,
466
+ history=None, return_history=False, generation_config=generation_config,
467
+ ))
468
+ thread.start()
469
+
470
+ # Initialize an empty string to store the generated text
471
+ generated_text = ''
472
+ # Loop through the streamer to get the new text as it is generated
473
+ for new_text in streamer:
474
+ if new_text == model.conv_template.sep:
475
+ break
476
+ generated_text += new_text
477
+ print(new_text, end='', flush=True) # Print each new chunk of generated text on the same line
478
+ ```
479
+
480
+ ## Finetune
481
+
482
+ Many repositories now support fine-tuning of the InternVL series models, including [InternVL](https://github.com/OpenGVLab/InternVL), [SWIFT](https://github.com/modelscope/ms-swift), [XTurner](https://github.com/InternLM/xtuner), and others. Please refer to their documentation for more details on fine-tuning.
483
+
484
+ ## Deployment
485
+
486
+ ### LMDeploy
487
+
488
+ LMDeploy is a toolkit for compressing, deploying, and serving LLMs & VLMs.
489
+
490
+ ```sh
491
+ pip install lmdeploy>=0.5.3
492
+ ```
493
+
494
+ LMDeploy abstracts the complex inference process of multi-modal Vision-Language Models (VLM) into an easy-to-use pipeline, similar to the Large Language Model (LLM) inference pipeline.
495
+
496
+ #### A 'Hello, world' Example
497
+
498
+ ```python
499
+ from lmdeploy import pipeline, TurbomindEngineConfig
500
+ from lmdeploy.vl import load_image
501
+
502
+ model = 'OpenGVLab/InternVL2-Llama3-76B'
503
+ image = load_image('https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/tests/data/tiger.jpeg')
504
+ pipe = pipeline(model, backend_config=TurbomindEngineConfig(session_len=8192, tp=4))
505
+ response = pipe(('describe this image', image))
506
+ print(response.text)
507
+ ```
508
+
509
+ If `ImportError` occurs while executing this case, please install the required dependency packages as prompted.
510
+
511
+ #### Multi-images Inference
512
+
513
+ When dealing with multiple images, you can put them all in one list. Keep in mind that multiple images will lead to a higher number of input tokens, and as a result, the size of the context window typically needs to be increased.
514
+
515
+ > Warning: Due to the scarcity of multi-image conversation data, the performance on multi-image tasks may be unstable, and it may require multiple attempts to achieve satisfactory results.
516
+
517
+ ```python
518
+ from lmdeploy import pipeline, TurbomindEngineConfig
519
+ from lmdeploy.vl import load_image
520
+ from lmdeploy.vl.constants import IMAGE_TOKEN
521
+
522
+ model = 'OpenGVLab/InternVL2-Llama3-76B'
523
+ pipe = pipeline(model, backend_config=TurbomindEngineConfig(session_len=8192, tp=4))
524
+
525
+ image_urls=[
526
+ 'https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/human-pose.jpg',
527
+ 'https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/det.jpg'
528
+ ]
529
+
530
+ images = [load_image(img_url) for img_url in image_urls]
531
+ # Numbering images improves multi-image conversations
532
+ response = pipe((f'Image-1: {IMAGE_TOKEN}\nImage-2: {IMAGE_TOKEN}\ndescribe these two images', images))
533
+ print(response.text)
534
+ ```
535
+
536
+ #### Batch Prompts Inference
537
+
538
+ Conducting inference with batch prompts is quite straightforward; just place them within a list structure:
539
+
540
+ ```python
541
+ from lmdeploy import pipeline, TurbomindEngineConfig
542
+ from lmdeploy.vl import load_image
543
+
544
+ model = 'OpenGVLab/InternVL2-Llama3-76B'
545
+ pipe = pipeline(model, backend_config=TurbomindEngineConfig(session_len=8192, tp=4))
546
+
547
+ image_urls=[
548
+ ""https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/human-pose.jpg"",
549
+ ""https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/det.jpg""
550
+ ]
551
+ prompts = [('describe this image', load_image(img_url)) for img_url in image_urls]
552
+ response = pipe(prompts)
553
+ print(response)
554
+ ```
555
+
556
+ #### Multi-turn Conversation
557
+
558
+ There are two ways to do the multi-turn conversations with the pipeline. One is to construct messages according to the format of OpenAI and use above introduced method, the other is to use the `pipeline.chat` interface.
559
+
560
+ ```python
561
+ from lmdeploy import pipeline, TurbomindEngineConfig, GenerationConfig
562
+ from lmdeploy.vl import load_image
563
+
564
+ model = 'OpenGVLab/InternVL2-Llama3-76B'
565
+ pipe = pipeline(model, backend_config=TurbomindEngineConfig(session_len=8192, tp=4))
566
+
567
+ image = load_image('https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/human-pose.jpg')
568
+ gen_config = GenerationConfig(top_k=40, top_p=0.8, temperature=0.8)
569
+ sess = pipe.chat(('describe this image', image), gen_config=gen_config)
570
+ print(sess.response.text)
571
+ sess = pipe.chat('What is the woman doing?', session=sess, gen_config=gen_config)
572
+ print(sess.response.text)
573
+ ```
574
+
575
+ #### Service
576
+
577
+ LMDeploy's `api_server` enables models to be easily packed into services with a single command. The provided RESTful APIs are compatible with OpenAI's interfaces. Below are an example of service startup:
578
+
579
+ ```shell
580
+ lmdeploy serve api_server OpenGVLab/InternVL2-Llama3-76B --server-port 23333 --tp 4
581
+ ```
582
+
583
+ To use the OpenAI-style interface, you need to install OpenAI:
584
+
585
+ ```shell
586
+ pip install openai
587
+ ```
588
+
589
+ Then, use the code below to make the API call:
590
+
591
+ ```python
592
+ from openai import OpenAI
593
+
594
+ client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:23333/v1')
595
+ model_name = client.models.list().data[0].id
596
+ response = client.chat.completions.create(
597
+ model=model_name,
598
+ messages=[{
599
+ 'role':
600
+ 'user',
601
+ 'content': [{
602
+ 'type': 'text',
603
+ 'text': 'describe this image',
604
+ }, {
605
+ 'type': 'image_url',
606
+ 'image_url': {
607
+ 'url':
608
+ 'https://modelscope.oss-cn-beijing.aliyuncs.com/resource/tiger.jpeg',
609
+ },
610
+ }],
611
+ }],
612
+ temperature=0.8,
613
+ top_p=0.8)
614
+ print(response)
615
+ ```
616
+
617
+ ## License
618
+
619
+ This project is released under the MIT License. This project uses the pre-trained Hermes-2-Theta-Llama-3-70B as a component, which is licensed under the Llama 3 Community License.
620
+
621
+ ## Citation
622
+
623
+ If you find this project useful in your research, please consider citing:
624
+
625
+ ```BibTeX
626
+ @article{chen2024expanding,
627
+ title={Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling},
628
+ author={Chen, Zhe and Wang, Weiyun and Cao, Yue and Liu, Yangzhou and Gao, Zhangwei and Cui, Erfei and Zhu, Jinguo and Ye, Shenglong and Tian, Hao and Liu, Zhaoyang and others},
629
+ journal={arXiv preprint arXiv:2412.05271},
630
+ year={2024}
631
+ }
632
+ @article{gao2024mini,
633
+ title={Mini-internvl: A flexible-transfer pocket multimodal model with 5\% parameters and 90\% performance},
634
+ author={Gao, Zhangwei and Chen, Zhe and Cui, Erfei and Ren, Yiming and Wang, Weiyun and Zhu, Jinguo and Tian, Hao and Ye, Shenglong and He, Junjun and Zhu, Xizhou and others},
635
+ journal={arXiv preprint arXiv:2410.16261},
636
+ year={2024}
637
+ }
638
+ @article{chen2024far,
639
+ title={How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suites},
640
+ author={Chen, Zhe and Wang, Weiyun and Tian, Hao and Ye, Shenglong and Gao, Zhangwei and Cui, Erfei and Tong, Wenwen and Hu, Kongzhi and Luo, Jiapeng and Ma, Zheng and others},
641
+ journal={arXiv preprint arXiv:2404.16821},
642
+ year={2024}
643
+ }
644
+ @inproceedings{chen2024internvl,
645
+ title={Internvl: Scaling up vision foundation models and aligning for generic visual-linguistic tasks},
646
+ author={Chen, Zhe and Wu, Jiannan and Wang, Wenhai and Su, Weijie and Chen, Guo and Xing, Sen and Zhong, Muyan and Zhang, Qinglong and Zhu, Xizhou and Lu, Lewei and others},
647
+ booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
648
+ pages={24185--24198},
649
+ year={2024}
650
+ }
651
+ ```
652
+ ","{""id"": ""OpenGVLab/InternVL2-Llama3-76B"", ""author"": ""OpenGVLab"", ""sha"": ""9d9aa88d8b76d0f38d6c30e15adbd26b5ceb1753"", ""last_modified"": ""2025-03-25 06:02:44+00:00"", ""created_at"": ""2024-07-15 06:16:18+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 399, ""downloads_all_time"": null, ""likes"": 213, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""safetensors"", ""internvl_chat"", ""feature-extraction"", ""internvl"", ""custom_code"", ""image-text-to-text"", ""conversational"", ""multilingual"", ""arxiv:2312.14238"", ""arxiv:2404.16821"", ""arxiv:2410.16261"", ""arxiv:2412.05271"", ""base_model:NousResearch/Hermes-2-Theta-Llama-3-70B"", ""base_model:merge:NousResearch/Hermes-2-Theta-Llama-3-70B"", ""base_model:OpenGVLab/InternViT-6B-448px-V1-5"", ""base_model:merge:OpenGVLab/InternViT-6B-448px-V1-5"", ""license:llama3"", ""region:us""], ""pipeline_tag"": ""image-text-to-text"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- OpenGVLab/InternViT-6B-448px-V1-5\n- NousResearch/Hermes-2-Theta-Llama-3-70B\nlanguage:\n- multilingual\nlibrary_name: transformers\nlicense: llama3\npipeline_tag: image-text-to-text\ntags:\n- internvl\n- custom_code\nnew_version: OpenGVLab/InternVL2_5-78B\nbase_model_relation: merge"", ""widget_data"": null, ""model_index"": null, ""config"": {""architectures"": [""InternVLChatModel""], ""auto_map"": {""AutoConfig"": ""configuration_internvl_chat.InternVLChatConfig"", ""AutoModel"": ""modeling_internvl_chat.InternVLChatModel"", ""AutoModelForCausalLM"": ""modeling_internvl_chat.InternVLChatModel""}, ""model_type"": ""internvl_chat"", ""tokenizer_config"": {""bos_token"": ""<|begin_of_text|>"", ""chat_template"": [{""name"": ""default"", ""template"": ""{{bos_token}}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}""}, {""name"": ""tool_use"", ""template"": ""{%- macro json_to_python_type(json_spec) %}\n{%- set basic_type_map = {\n \""string\"": \""str\"",\n \""number\"": \""float\"",\n \""integer\"": \""int\"",\n \""boolean\"": \""bool\""\n} %}\n\n{%- if basic_type_map[json_spec.type] is defined %}\n {{- basic_type_map[json_spec.type] }}\n{%- elif json_spec.type == \""array\"" %}\n {{- \""list[\"" + json_to_python_type(json_spec|items) + \""]\""}}\n{%- elif json_spec.type == \""object\"" %}\n {%- if json_spec.additionalProperties is defined %}\n {{- \""dict[str, \"" + json_to_python_type(json_spec.additionalProperties) + ']'}}\n {%- else %}\n {{- \""dict\"" }}\n {%- endif %}\n{%- elif json_spec.type is iterable %}\n {{- \""Union[\"" }}\n {%- for t in json_spec.type %}\n {{- json_to_python_type({\""type\"": t}) }}\n {%- if not loop.last %}\n {{- \"",\"" }} \n {%- endif %}\n {%- endfor %}\n {{- \""]\"" }}\n{%- else %}\n {{- \""Any\"" }}\n{%- endif %}\n{%- endmacro %}\n\n\n{{- bos_token }}\n{{- \""You are a function calling AI model. You are provided with function signatures within <tools></tools> XML tags. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions. Here are the available tools: <tools> \"" }}\n{%- for tool in tools %}\n {%- if tool.function is defined %}\n {%- set tool = tool.function %}\n {%- endif %}\n {{- '{\""type\"": \""function\"", \""function\"": ' }}\n {{- '{\""name\"": ' + tool.name + '\"", ' }}\n {{- '\""description\"": \""' + tool.name + '(' }}\n {%- for param_name, param_fields in tool.parameters.properties|items %}\n {{- param_name + \"": \"" + json_to_python_type(param_fields) }}\n {%- if not loop.last %}\n {{- \"", \"" }}\n {%- endif %}\n {%- endfor %}\n {{- \"")\"" }}\n {%- if tool.return is defined %}\n {{- \"" -> \"" + json_to_python_type(tool.return) }}\n {%- endif %}\n {{- \"" - \"" + tool.description + \""\\n\\n\"" }}\n {%- for param_name, param_fields in tool.parameters.properties|items %}\n {%- if loop.first %}\n {{- \"" Args:\\n\"" }}\n {%- endif %}\n {{- \"" \"" + param_name + \""(\"" + json_to_python_type(param_fields) + \""): \"" + param_fields.description|trim }}\n {%- endfor %}\n {%- if tool.return is defined and tool.return.description is defined %}\n {{- \""\\n Returns:\\n \"" + tool.return.description }}\n {%- endif %}\n {{- '\""' }}\n {{- ', \""parameters\"": ' }}\n {%- if tool.parameters.properties | length == 0 %}\n {{- \""{}\"" }}\n {%- else %}\n {{- tool.parameters|tojson }}\n {%- endif %}\n {{- \""}\"" }}\n {%- if not loop.last %}\n {{- \""\\n\"" }}\n {%- endif %}\n{%- endfor %}\n{{- \"" </tools>\"" }}\n{{- 'Use the following pydantic model json schema for each tool call you will make: {\""properties\"": {\""arguments\"": {\""title\"": \""Arguments\"", \""type\"": \""object\""}, \""name\"": {\""title\"": \""Name\"", \""type\"": \""string\""}}, \""required\"": [\""arguments\"", \""name\""], \""title\"": \""FunctionCall\"", \""type\"": \""object\""}\n' }}\n{{- \""For each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:\n\"" }}\n{{- \""<tool_call>\n\"" }}\n{{- '{\""arguments\"": <args-dict>, \""name\"": <function-name>}\n' }}\n{{- '</tool_call><|im_end|>' }}\n{%- for message in messages %}\n {%- if message.role == \""user\"" or message.role == \""system\"" or (message.role == \""assistant\"" and message.tool_calls is not defined) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \""assistant\"" %}\n {{- '<|im_start|>' + message.role + '\\n<tool_call>\\n' }}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '{ ' }}\n {%- if tool_call.arguments is defined %}\n {{- '\""arguments\"": ' }}\n {{- tool_call.arguments|tojson }}\n {{- ', '}}\n {%- endif %}\n {{- '\""name\"": \""' }}\n {{- tool_call.name }}\n {{- '\""}' }}\n {{- '\\n</tool_call> ' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \""tool\"" %}\n {%- if not message.name is defined %}\n {{- raise_exception(\""Tool response dicts require a 'name' key indicating the name of the called function!\"") }}\n {%- endif %}\n {{- '<|im_start|>' + message.role + '\\n<tool_response>\\n' }}\n {{- '{\""name\"": \""' }}\n {{- message.name }}\n {{- '\"", \""content\"": ' }}\n {{- message.content|tojson + '}' }}\n {{- '\\n</tool_response> <|im_end|>\\n' }} \n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n""}], ""eos_token"": ""<|im_end|>"", ""pad_token"": ""<|end_of_text|>""}}, ""transformers_info"": {""auto_model"": ""AutoModel"", ""custom_class"": ""modeling_internvl_chat.InternVLChatModel"", ""pipeline_tag"": ""feature-extraction"", ""processor"": null}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='configuration_intern_vit.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='configuration_internvl_chat.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='conversation.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='examples/image1.jpg', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='examples/image2.jpg', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='examples/red-panda.mp4', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00001-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00002-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00003-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00004-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00005-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00006-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00007-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00008-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00009-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00010-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00011-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00012-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00013-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00014-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00015-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00016-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00017-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00018-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00019-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00020-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00021-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00022-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00023-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00024-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00025-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00026-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00027-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00028-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00029-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00030-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00031-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00032-of-00032.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='modeling_intern_vit.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='modeling_internvl_chat.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [""TIGER-Lab/MEGA-Bench"", ""awacke1/Leaderboard-Deepseek-Gemini-Grok-GPT-Qwen"", ""torettomarui/Llava-qw""], ""safetensors"": {""parameters"": {""BF16"": 76262358400}, ""total"": 76262358400}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-03-25 06:02:44+00:00"", ""cardData"": ""base_model:\n- OpenGVLab/InternViT-6B-448px-V1-5\n- NousResearch/Hermes-2-Theta-Llama-3-70B\nlanguage:\n- multilingual\nlibrary_name: transformers\nlicense: llama3\npipeline_tag: image-text-to-text\ntags:\n- internvl\n- custom_code\nnew_version: OpenGVLab/InternVL2_5-78B\nbase_model_relation: merge"", ""transformersInfo"": {""auto_model"": ""AutoModel"", ""custom_class"": ""modeling_internvl_chat.InternVLChatModel"", ""pipeline_tag"": ""feature-extraction"", ""processor"": null}, ""_id"": ""6694beb2a5108c33e684dd62"", ""modelId"": ""OpenGVLab/InternVL2-Llama3-76B"", ""usedStorage"": 152526875869}",0,,0,,0,https://huggingface.co/OpenGVLab/InternVL2-Llama3-76B-AWQ,1,,0,"OpenGVLab/InternVL, TIGER-Lab/MEGA-Bench, awacke1/Leaderboard-Deepseek-Gemini-Grok-GPT-Qwen, huggingface/InferenceSupport/discussions/new?title=OpenGVLab/InternVL2-Llama3-76B&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BOpenGVLab%2FInternVL2-Llama3-76B%5D(%2FOpenGVLab%2FInternVL2-Llama3-76B)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, torettomarui/Llava-qw",5
Janus-Pro-1B_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv ADDED
@@ -0,0 +1,120 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ deepseek-ai/Janus-Pro-1B,"---
3
+ license: mit
4
+ license_name: deepseek
5
+ license_link: LICENSE
6
+ pipeline_tag: any-to-any
7
+ library_name: transformers
8
+ tags:
9
+ - muiltimodal
10
+ - text-to-image
11
+ - unified-model
12
+ ---
13
+
14
+ ## 1. Introduction
15
+
16
+ Janus-Pro is a novel autoregressive framework that unifies multimodal understanding and generation.
17
+ It addresses the limitations of previous approaches by decoupling visual encoding into separate pathways, while still utilizing a single, unified transformer architecture for processing. The decoupling not only alleviates the conflict between the visual encoder’s roles in understanding and generation, but also enhances the framework’s flexibility.
18
+ Janus-Pro surpasses previous unified model and matches or exceeds the performance of task-specific models.
19
+ The simplicity, high flexibility, and effectiveness of Janus-Pro make it a strong candidate for next-generation unified multimodal models.
20
+
21
+ [**Github Repository**](https://github.com/deepseek-ai/Janus)
22
+
23
+ <div align=""center"">
24
+ <img alt=""image"" src=""janus_pro_teaser1.png"" style=""width:90%;"">
25
+ </div>
26
+
27
+ <div align=""center"">
28
+ <img alt=""image"" src=""janus_pro_teaser2.png"" style=""width:90%;"">
29
+ </div>
30
+
31
+
32
+ ### 2. Model Summary
33
+
34
+ Janus-Pro is a unified understanding and generation MLLM, which decouples visual encoding for multimodal understanding and generation.
35
+ Janus-Pro is constructed based on the DeepSeek-LLM-1.5b-base/DeepSeek-LLM-7b-base.
36
+
37
+ For multimodal understanding, it uses the [SigLIP-L](https://huggingface.co/timm/ViT-L-16-SigLIP-384) as the vision encoder, which supports 384 x 384 image input. For image generation, Janus-Pro uses the tokenizer from [here](https://github.com/FoundationVision/LlamaGen) with a downsample rate of 16.
38
+
39
+
40
+
41
+ ## 3. Quick Start
42
+
43
+ Please refer to [**Github Repository**](https://github.com/deepseek-ai/Janus)
44
+
45
+
46
+ ## 4. License
47
+
48
+ This code repository is licensed under [the MIT License](https://github.com/deepseek-ai/DeepSeek-LLM/blob/HEAD/LICENSE-CODE). The use of Janus-Pro models is subject to [DeepSeek Model License](https://github.com/deepseek-ai/DeepSeek-LLM/blob/HEAD/LICENSE-MODEL).
49
+ ## 5. Citation
50
+
51
+ ```
52
+ @article{chen2025janus,
53
+ title={Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling},
54
+ author={Chen, Xiaokang and Wu, Zhiyu and Liu, Xingchao and Pan, Zizheng and Liu, Wen and Xie, Zhenda and Yu, Xingkai and Ruan, Chong},
55
+ journal={arXiv preprint arXiv:2501.17811},
56
+ year={2025}
57
+ }
58
+ ```
59
+
60
+ ## 6. Contact
61
+
62
+ If you have any questions, please raise an issue or contact us at [service@deepseek.com](mailto:service@deepseek.com).","{""id"": ""deepseek-ai/Janus-Pro-1B"", ""author"": ""deepseek-ai"", ""sha"": ""960ab33191f61342a4c60ae74d8dc356a39fafcb"", ""last_modified"": ""2025-02-01 08:00:32+00:00"", ""created_at"": ""2025-01-26 12:07:49+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 34808, ""downloads_all_time"": null, ""likes"": 432, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""pytorch"", ""multi_modality"", ""muiltimodal"", ""text-to-image"", ""unified-model"", ""any-to-any"", ""arxiv:2501.17811"", ""license:mit"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""any-to-any"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""library_name: transformers\nlicense: mit\nlicense_name: deepseek\nlicense_link: LICENSE\npipeline_tag: any-to-any\ntags:\n- muiltimodal\n- text-to-image\n- unified-model"", ""widget_data"": null, ""model_index"": null, ""config"": {""architectures"": [""MultiModalityCausalLM""], ""model_type"": ""multi_modality"", ""tokenizer_config"": {""bos_token"": ""<\uff5cbegin\u2581of\u2581sentence\uff5c>"", ""eos_token"": ""<\uff5cend\u2581of\u2581sentence\uff5c>"", ""pad_token"": null, ""unk_token"": null, ""use_default_system_prompt"": true}}, ""transformers_info"": {""auto_model"": ""MultiModalityCausalLM"", ""custom_class"": null, ""pipeline_tag"": null, ""processor"": null}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='janus_pro_teaser1.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='janus_pro_teaser2.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='processor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [""afrideva/Janus-Pro-1b"", ""mgbam/image"", ""Gemini899/Janus-Pro-1b"", ""techmilano/cloud-vector-ai-deekseek-janus"", ""techmilano/cvai-deepseek-janus"", ""NomiDecent/deepseek_Multimodal_RAG""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-01 08:00:32+00:00"", ""cardData"": ""library_name: transformers\nlicense: mit\nlicense_name: deepseek\nlicense_link: LICENSE\npipeline_tag: any-to-any\ntags:\n- muiltimodal\n- text-to-image\n- unified-model"", ""transformersInfo"": {""auto_model"": ""MultiModalityCausalLM"", ""custom_class"": null, ""pipeline_tag"": null, ""processor"": null}, ""_id"": ""67962595a08d7b966abd37e9"", ""modelId"": ""deepseek-ai/Janus-Pro-1B"", ""usedStorage"": 8357757117}",0,"https://huggingface.co/wnma3mz/Janus-Pro-1B-LM, https://huggingface.co/wnma3mz/Janus-Pro-1B",2,,0,"https://huggingface.co/onnx-community/Janus-Pro-1B-ONNX, https://huggingface.co/wnma3mz/Janus-Pro-1B-4bit, https://huggingface.co/Casalioy/Janus-Pro-1B-q4-Casa_MLX, https://huggingface.co/AXERA-TECH/Janus-Pro-1B, https://huggingface.co/wnma3mz/Janus-Pro-1B-LM-4bit",5,,0,"Gemini899/Janus-Pro-1b, NomiDecent/deepseek_Multimodal_RAG, afrideva/Janus-Pro-1b, huggingface/InferenceSupport/discussions/new?title=deepseek-ai/Janus-Pro-1B&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bdeepseek-ai%2FJanus-Pro-1B%5D(%2Fdeepseek-ai%2FJanus-Pro-1B)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, mgbam/image, techmilano/cloud-vector-ai-deekseek-janus, techmilano/cvai-deepseek-janus",7
63
+ wnma3mz/Janus-Pro-1B-LM,"---
64
+ license: mit
65
+ license_name: deepseek
66
+ license_link: LICENSE
67
+ pipeline_tag: text-generation
68
+ library_name: transformers
69
+ base_model:
70
+ - deepseek-ai/Janus-Pro-1B
71
+ tags:
72
+ - chat
73
+ ---
74
+
75
+ This model is derived from https://huggingface.co/deepseek-ai/Janus-Pro-1B and the main modifications are as follows
76
+
77
+ - bin files are updated to safetensors
78
+ - Add chat_template
79
+
80
+ `4bit` refers to quantifying the LLM part to 4 bits.
81
+
82
+ `LM` means that it contains only the language model part.
83
+
84
+ ## Quick Start
85
+
86
+ In Macos (Apple silicon), use [mlx](https://github.com/ml-explore/mlx) framework https://github.com/wnma3mz/tLLM
87
+
88
+ ```bash
89
+ tllm.server --model_path $MODEL_PATH --hostname localhost --is_local --client_size 1
90
+ ```
91
+
92
+ `$MODEL_PATH` like `wnma3mz/Janus-Pro-1B-4bit`","{""id"": ""wnma3mz/Janus-Pro-1B-LM"", ""author"": ""wnma3mz"", ""sha"": ""d993cf5e0c08d5b9909f57a1c0e34a1f541b9917"", ""last_modified"": ""2025-01-30 05:34:33+00:00"", ""created_at"": ""2025-01-28 12:22:22+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 15, ""downloads_all_time"": null, ""likes"": 2, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""safetensors"", ""llama"", ""text-generation"", ""chat"", ""conversational"", ""base_model:deepseek-ai/Janus-Pro-1B"", ""base_model:finetune:deepseek-ai/Janus-Pro-1B"", ""license:mit"", ""autotrain_compatible"", ""text-generation-inference"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- deepseek-ai/Janus-Pro-1B\nlibrary_name: transformers\nlicense: mit\nlicense_name: deepseek\nlicense_link: LICENSE\npipeline_tag: text-generation\ntags:\n- chat"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": null, ""config"": {""architectures"": [""LlamaForCausalLM""], ""model_type"": ""llama"", ""tokenizer_config"": {""bos_token"": ""<\uff5cbegin\u2581of\u2581sentence\uff5c>"", ""eos_token"": ""<\uff5cend\u2581of\u2581sentence\uff5c>"", ""pad_token"": null, ""unk_token"": null, ""chat_template"": ""{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set ns = namespace(is_first=false, is_tool=false, is_output_first=true, system_prompt='') %}{%- for message in messages %}{%- if message['role'] == 'system' %}{% set ns.system_prompt = message['content'] %}{%- endif %}{%- endfor %}{{ns.system_prompt}}{%- for message in messages %}{%- if message['role'] == 'user' %}{{'<|User|>: ' + message['content'] + '\\n\\n'}}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is not none %}{% set content = message['content'] %}{% if '</think>' in content %}{% set content = content.split('</think>')[-1] %}{% endif %}{{'<|Assistant|>: ' + content + '<\uff5cend\u2581of\u2581sentence\uff5c>'}}{%- endif %}{%- endfor -%}{% if add_generation_prompt %}{{'<|Assistant|>: '}}{% endif %}"", ""use_default_system_prompt"": true}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='janus_pro_teaser1.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='janus_pro_teaser2.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='processor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": {""parameters"": {""F16"": 1652656128}, ""total"": 1652656128}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-01-30 05:34:33+00:00"", ""cardData"": ""base_model:\n- deepseek-ai/Janus-Pro-1B\nlibrary_name: transformers\nlicense: mit\nlicense_name: deepseek\nlicense_link: LICENSE\npipeline_tag: text-generation\ntags:\n- chat"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""_id"": ""6798cbfe4aa0c8099f164f5d"", ""modelId"": ""wnma3mz/Janus-Pro-1B-LM"", ""usedStorage"": 3305337400}",1,,0,,0,"https://huggingface.co/mradermacher/Janus-Pro-1B-LM-i1-GGUF, https://huggingface.co/mradermacher/Janus-Pro-1B-LM-GGUF",2,,0,huggingface/InferenceSupport/discussions/new?title=wnma3mz/Janus-Pro-1B-LM&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bwnma3mz%2FJanus-Pro-1B-LM%5D(%2Fwnma3mz%2FJanus-Pro-1B-LM)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
93
+ wnma3mz/Janus-Pro-1B,"---
94
+ license: mit
95
+ license_name: deepseek
96
+ license_link: LICENSE
97
+ pipeline_tag: any-to-any
98
+ library_name: mlx
99
+ base_model:
100
+ - deepseek-ai/Janus-Pro-1B
101
+ tags:
102
+ - chat
103
+ ---
104
+
105
+ This model is derived from https://huggingface.co/deepseek-ai/Janus-Pro-1B and the main modifications are as follows
106
+
107
+ - bin files are updated to safetensors
108
+ - Add chat_template
109
+
110
+ `4bit` mainly refers to quantifying the LLM part to 4 bits.
111
+
112
+ ## Quick Start
113
+
114
+ In Macos (Apple silicon), use [mlx](https://github.com/ml-explore/mlx) framework https://github.com/wnma3mz/tLLM
115
+
116
+ ```bash
117
+ tllm.server --model_path $MODEL_PATH
118
+ ```
119
+
120
+ `$MODEL_PATH` like `wnma3mz/Janus-Pro-1B-4bit`","{""id"": ""wnma3mz/Janus-Pro-1B"", ""author"": ""wnma3mz"", ""sha"": ""0fae24a328de292cd74d092a6898914a49d07cec"", ""last_modified"": ""2025-02-01 13:00:33+00:00"", ""created_at"": ""2025-01-30 05:45:19+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 21, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""mlx"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""mlx"", ""safetensors"", ""chat"", ""any-to-any"", ""base_model:deepseek-ai/Janus-Pro-1B"", ""base_model:finetune:deepseek-ai/Janus-Pro-1B"", ""license:mit"", ""region:us""], ""pipeline_tag"": ""any-to-any"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- deepseek-ai/Janus-Pro-1B\nlibrary_name: mlx\nlicense: mit\nlicense_name: deepseek\nlicense_link: LICENSE\npipeline_tag: any-to-any\ntags:\n- chat"", ""widget_data"": null, ""model_index"": null, ""config"": {""architectures"": [""JanusProConditionalGeneration""], ""tokenizer_config"": {""bos_token"": ""<\uff5cbegin\u2581of\u2581sentence\uff5c>"", ""eos_token"": ""<\uff5cend\u2581of\u2581sentence\uff5c>"", ""pad_token"": null, ""unk_token"": null, ""chat_template"": ""{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set ns = namespace(is_first=false, is_tool=false, is_output_first=true, system_prompt='You are a helpful language and vision assistant. You are able to understand the visual content that the user provides, and assist the user with a variety of tasks using natural language.') %}{%- for message in messages %}{%- if message['role'] == 'system' %}{% set ns.system_prompt = message['content'] %}{% break %}{%- endif %}{%- endfor %}{{ns.system_prompt}}{% if ns.system_prompt %}{{'\n\n'}}{% endif %}{%- for message in messages %}{%- if message['role'] == 'user' %}{{'<|User|>: ' + message['content'] + '\n\n'}}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is not none %}{% set content = message['content'] %}{% if '</think>' in content %}{% set content = content.split('</think>')[-1] %}{% endif %}{{'<|Assistant|>: ' + content}}{% if add_generation_prompt or '</think>' in content %}{{'<\uff5cend of sentence\uff5c>'}}{% endif %}{%- endif %}{%- endfor -%}{% if add_generation_prompt %}{{'<|Assistant|>:'}}{% endif %}"", ""use_default_system_prompt"": true}}, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='processor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": {""parameters"": {""F16"": 2089297547}, ""total"": 2089297547}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-01 13:00:33+00:00"", ""cardData"": ""base_model:\n- deepseek-ai/Janus-Pro-1B\nlibrary_name: mlx\nlicense: mit\nlicense_name: deepseek\nlicense_link: LICENSE\npipeline_tag: any-to-any\ntags:\n- chat"", ""transformersInfo"": null, ""_id"": ""679b11ef25cd3e855df85ff0"", ""modelId"": ""wnma3mz/Janus-Pro-1B"", ""usedStorage"": 4178705502}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=wnma3mz/Janus-Pro-1B&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bwnma3mz%2FJanus-Pro-1B%5D(%2Fwnma3mz%2FJanus-Pro-1B)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
Janus-Pro-7B_finetunes_20250422_225821.csv ADDED
The diff for this file is too large to render. See raw diff
 
LLaMA-7B_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ nyanko7/LLaMA-7B,"---
3
+ license: openrail
4
+ ---
5
+ ","{""id"": ""nyanko7/LLaMA-7B"", ""author"": ""nyanko7"", ""sha"": ""3b94aad3d8e39981b81e14239446d935b6dbf2bd"", ""last_modified"": ""2023-03-04 02:18:23+00:00"", ""created_at"": ""2023-03-04 01:59:47+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 211, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""license:openrail"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""license: openrail"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='checklist.chk', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='consolidated.00.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='params.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.model', size=None, blob_id=None, lfs=None)""], ""spaces"": [""csuhan/LLaMA-Adapter"", ""xuxw98/TAPA"", ""heikowagner/GPT-Docker"", ""heikowagner/GPT-Docker2"", ""UTOPIA-DXD/LLaMA-Adapter""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-03-04 02:18:23+00:00"", ""cardData"": ""license: openrail"", ""transformersInfo"": null, ""_id"": ""6402a61302594ec43ec340a4"", ""modelId"": ""nyanko7/LLaMA-7B"", ""usedStorage"": 13477439239}",0,https://huggingface.co/haarismian/cyberbullying-llama,1,,0,,0,,0,"UTOPIA-DXD/LLaMA-Adapter, csuhan/LLaMA-Adapter, heikowagner/GPT-Docker, heikowagner/GPT-Docker2, huggingface/InferenceSupport/discussions/new?title=nyanko7/LLaMA-7B&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bnyanko7%2FLLaMA-7B%5D(%2Fnyanko7%2FLLaMA-7B)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, xuxw98/TAPA",6
6
+ haarismian/cyberbullying-llama,"---
7
+ license: mit
8
+ language:
9
+ - en
10
+ base_model:
11
+ - nyanko7/LLaMA-7B
12
+ datasets:
13
+ - haarismian/ejaz-et-al-dataset
14
+ metrics:
15
+ - accuracy
16
+ ---
17
+ # Model Card for Model ID
18
+
19
+ <!-- Provide a quick summary of what the model is/does. -->
20
+
21
+ This modelcard aims to be a base template for new models. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/modelcard_template.md?plain=1).
22
+
23
+ ## Model Details
24
+
25
+ ### Model Description
26
+
27
+ <!-- Provide a longer summary of what this model is. -->
28
+
29
+
30
+
31
+ - **Developed by:** Haaris Mian
32
+ - **Funded by [optional]:** Self Funded for MSc Dissertation
33
+
34
+ - **Finetuned from model [optional]:** LLaMA 2 7B
35
+
36
+ ### Model Sources [optional]
37
+
38
+ <!-- Provide the basic links for the model. -->
39
+
40
+ - **Repository:** [More Information Needed]
41
+ - **Paper [optional]:** [More Information Needed]
42
+ - **Demo [optional]:** [More Information Needed]
43
+
44
+ ## Uses
45
+
46
+ <!-- The model is intended for research and academic purposes in understanding and mitigating cyberbullying. It can be used to detect harmful online interactions and assist in content moderation.-->
47
+
48
+ ### Direct Use
49
+
50
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
51
+
52
+ [More Information Needed]
53
+
54
+ ### Downstream Use [optional]
55
+
56
+ <!-- Automated moderation of social media platforms.
57
+
58
+ Enhancing AI-powered chatbot moderation.
59
+
60
+ Supporting online safety tools and AI-driven reporting mechanisms. -->","{""id"": ""haarismian/cyberbullying-llama"", ""author"": ""haarismian"", ""sha"": ""467bea92c38ada2f8d14fde65856d92e2e76ac9b"", ""last_modified"": ""2025-02-24 00:38:02+00:00"", ""created_at"": ""2025-02-24 00:15:24+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""en"", ""dataset:haarismian/ejaz-et-al-dataset"", ""base_model:nyanko7/LLaMA-7B"", ""base_model:finetune:nyanko7/LLaMA-7B"", ""license:mit"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- nyanko7/LLaMA-7B\ndatasets:\n- haarismian/ejaz-et-al-dataset\nlanguage:\n- en\nlicense: mit\nmetrics:\n- accuracy"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-24 00:38:02+00:00"", ""cardData"": ""base_model:\n- nyanko7/LLaMA-7B\ndatasets:\n- haarismian/ejaz-et-al-dataset\nlanguage:\n- en\nlicense: mit\nmetrics:\n- accuracy"", ""transformersInfo"": null, ""_id"": ""67bbba1c87fb28f3ecc01e4d"", ""modelId"": ""haarismian/cyberbullying-llama"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=haarismian/cyberbullying-llama&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bhaarismian%2Fcyberbullying-llama%5D(%2Fhaarismian%2Fcyberbullying-llama)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
Llama-2-7b-hf_finetunes_20250427_003734.csv_finetunes_20250427_003734.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama-3-8B-Instruct-262k_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama-3-8B-Lexi-Uncensored_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv ADDED
@@ -0,0 +1,273 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ Orenguteng/Llama-3-8B-Lexi-Uncensored,"---
3
+ license: llama3
4
+ tags:
5
+ - uncensored
6
+ - llama3
7
+ - instruct
8
+ - open
9
+ model-index:
10
+ - name: Llama-3-8B-Lexi-Uncensored
11
+ results:
12
+ - task:
13
+ type: text-generation
14
+ name: Text Generation
15
+ dataset:
16
+ name: AI2 Reasoning Challenge (25-Shot)
17
+ type: ai2_arc
18
+ config: ARC-Challenge
19
+ split: test
20
+ args:
21
+ num_few_shot: 25
22
+ metrics:
23
+ - type: acc_norm
24
+ value: 59.56
25
+ name: normalized accuracy
26
+ source:
27
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored
28
+ name: Open LLM Leaderboard
29
+ - task:
30
+ type: text-generation
31
+ name: Text Generation
32
+ dataset:
33
+ name: HellaSwag (10-Shot)
34
+ type: hellaswag
35
+ split: validation
36
+ args:
37
+ num_few_shot: 10
38
+ metrics:
39
+ - type: acc_norm
40
+ value: 77.88
41
+ name: normalized accuracy
42
+ source:
43
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored
44
+ name: Open LLM Leaderboard
45
+ - task:
46
+ type: text-generation
47
+ name: Text Generation
48
+ dataset:
49
+ name: MMLU (5-Shot)
50
+ type: cais/mmlu
51
+ config: all
52
+ split: test
53
+ args:
54
+ num_few_shot: 5
55
+ metrics:
56
+ - type: acc
57
+ value: 67.68
58
+ name: accuracy
59
+ source:
60
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored
61
+ name: Open LLM Leaderboard
62
+ - task:
63
+ type: text-generation
64
+ name: Text Generation
65
+ dataset:
66
+ name: TruthfulQA (0-shot)
67
+ type: truthful_qa
68
+ config: multiple_choice
69
+ split: validation
70
+ args:
71
+ num_few_shot: 0
72
+ metrics:
73
+ - type: mc2
74
+ value: 47.72
75
+ source:
76
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored
77
+ name: Open LLM Leaderboard
78
+ - task:
79
+ type: text-generation
80
+ name: Text Generation
81
+ dataset:
82
+ name: Winogrande (5-shot)
83
+ type: winogrande
84
+ config: winogrande_xl
85
+ split: validation
86
+ args:
87
+ num_few_shot: 5
88
+ metrics:
89
+ - type: acc
90
+ value: 75.85
91
+ name: accuracy
92
+ source:
93
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored
94
+ name: Open LLM Leaderboard
95
+ - task:
96
+ type: text-generation
97
+ name: Text Generation
98
+ dataset:
99
+ name: GSM8k (5-shot)
100
+ type: gsm8k
101
+ config: main
102
+ split: test
103
+ args:
104
+ num_few_shot: 5
105
+ metrics:
106
+ - type: acc
107
+ value: 68.39
108
+ name: accuracy
109
+ source:
110
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored
111
+ name: Open LLM Leaderboard
112
+ ---
113
+
114
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/644ad182f434a6a63b18eee6/H6axm5mlmiOWnbIFvx_em.png)
115
+
116
+ This model is based on Llama-3-8b-Instruct, and is governed by [META LLAMA 3 COMMUNITY LICENSE AGREEMENT](https://llama.meta.com/llama3/license/)
117
+
118
+ Lexi is uncensored, which makes the model compliant. You are advised to implement your own alignment layer before exposing the model as a service. It will be highly compliant with any requests, even unethical ones.
119
+
120
+ You are responsible for any content you create using this model. Please use it responsibly.
121
+
122
+ Lexi is licensed according to Meta's Llama license. I grant permission for any use, including commercial, that falls within accordance with Meta's Llama-3 license.
123
+
124
+ # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
125
+ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Orenguteng__Llama-3-8B-Lexi-Uncensored)
126
+
127
+ | Metric |Value|
128
+ |---------------------------------|----:|
129
+ |Avg. |66.18|
130
+ |AI2 Reasoning Challenge (25-Shot)|59.56|
131
+ |HellaSwag (10-Shot) |77.88|
132
+ |MMLU (5-Shot) |67.68|
133
+ |TruthfulQA (0-shot) |47.72|
134
+ |Winogrande (5-shot) |75.85|
135
+ |GSM8k (5-shot) |68.39|
136
+
137
+ ","{""id"": ""Orenguteng/Llama-3-8B-Lexi-Uncensored"", ""author"": ""Orenguteng"", ""sha"": ""ff95e3bfcd6142759ce82099b58bc7a789ac241b"", ""last_modified"": ""2024-05-27 06:16:40+00:00"", ""created_at"": ""2024-04-23 21:14:40+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 284967, ""downloads_all_time"": null, ""likes"": 219, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""safetensors"", ""llama"", ""text-generation"", ""uncensored"", ""llama3"", ""instruct"", ""open"", ""conversational"", ""license:llama3"", ""model-index"", ""autotrain_compatible"", ""text-generation-inference"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""license: llama3\ntags:\n- uncensored\n- llama3\n- instruct\n- open\nmodel-index:\n- name: Llama-3-8B-Lexi-Uncensored\n results:\n - task:\n type: text-generation\n name: Text Generation\n dataset:\n name: AI2 Reasoning Challenge (25-Shot)\n type: ai2_arc\n config: ARC-Challenge\n split: test\n args:\n num_few_shot: 25\n metrics:\n - type: acc_norm\n value: 59.56\n name: normalized accuracy\n verified: false\n source:\n url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored\n name: Open LLM Leaderboard\n - task:\n type: text-generation\n name: Text Generation\n dataset:\n name: HellaSwag (10-Shot)\n type: hellaswag\n split: validation\n args:\n num_few_shot: 10\n metrics:\n - type: acc_norm\n value: 77.88\n name: normalized accuracy\n verified: false\n source:\n url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored\n name: Open LLM Leaderboard\n - task:\n type: text-generation\n name: Text Generation\n dataset:\n name: MMLU (5-Shot)\n type: cais/mmlu\n config: all\n split: test\n args:\n num_few_shot: 5\n metrics:\n - type: acc\n value: 67.68\n name: accuracy\n verified: false\n source:\n url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored\n name: Open LLM Leaderboard\n - task:\n type: text-generation\n name: Text Generation\n dataset:\n name: TruthfulQA (0-shot)\n type: truthful_qa\n config: multiple_choice\n split: validation\n args:\n num_few_shot: 0\n metrics:\n - type: mc2\n value: 47.72\n verified: false\n source:\n url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored\n name: Open LLM Leaderboard\n - task:\n type: text-generation\n name: Text Generation\n dataset:\n name: Winogrande (5-shot)\n type: winogrande\n config: winogrande_xl\n split: validation\n args:\n num_few_shot: 5\n metrics:\n - type: acc\n value: 75.85\n name: accuracy\n verified: false\n source:\n url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored\n name: Open LLM Leaderboard\n - task:\n type: text-generation\n name: Text Generation\n dataset:\n name: GSM8k (5-shot)\n type: gsm8k\n config: main\n split: test\n args:\n num_few_shot: 5\n metrics:\n - type: acc\n value: 68.39\n name: accuracy\n verified: false\n source:\n url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored\n name: Open LLM Leaderboard"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": [{""name"": ""Llama-3-8B-Lexi-Uncensored"", ""results"": [{""task"": {""type"": ""text-generation"", ""name"": ""Text Generation""}, ""dataset"": {""name"": ""AI2 Reasoning Challenge (25-Shot)"", ""type"": ""ai2_arc"", ""config"": ""ARC-Challenge"", ""split"": ""test"", ""args"": {""num_few_shot"": 25}}, ""metrics"": [{""type"": ""acc_norm"", ""value"": 59.56, ""name"": ""normalized accuracy"", ""verified"": false}], ""source"": {""url"": ""https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored"", ""name"": ""Open LLM Leaderboard""}}, {""task"": {""type"": ""text-generation"", ""name"": ""Text Generation""}, ""dataset"": {""name"": ""HellaSwag (10-Shot)"", ""type"": ""hellaswag"", ""split"": ""validation"", ""args"": {""num_few_shot"": 10}}, ""metrics"": [{""type"": ""acc_norm"", ""value"": 77.88, ""name"": ""normalized accuracy"", ""verified"": false}], ""source"": {""url"": ""https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored"", ""name"": ""Open LLM Leaderboard""}}, {""task"": {""type"": ""text-generation"", ""name"": ""Text Generation""}, ""dataset"": {""name"": ""MMLU (5-Shot)"", ""type"": ""cais/mmlu"", ""config"": ""all"", ""split"": ""test"", ""args"": {""num_few_shot"": 5}}, ""metrics"": [{""type"": ""acc"", ""value"": 67.68, ""name"": ""accuracy"", ""verified"": false}], ""source"": {""url"": ""https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored"", ""name"": ""Open LLM Leaderboard""}}, {""task"": {""type"": ""text-generation"", ""name"": ""Text Generation""}, ""dataset"": {""name"": ""TruthfulQA (0-shot)"", ""type"": ""truthful_qa"", ""config"": ""multiple_choice"", ""split"": ""validation"", ""args"": {""num_few_shot"": 0}}, ""metrics"": [{""type"": ""mc2"", ""value"": 47.72, ""verified"": false}], ""source"": {""url"": ""https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored"", ""name"": ""Open LLM Leaderboard""}}, {""task"": {""type"": ""text-generation"", ""name"": ""Text Generation""}, ""dataset"": {""name"": ""Winogrande (5-shot)"", ""type"": ""winogrande"", ""config"": ""winogrande_xl"", ""split"": ""validation"", ""args"": {""num_few_shot"": 5}}, ""metrics"": [{""type"": ""acc"", ""value"": 75.85, ""name"": ""accuracy"", ""verified"": false}], ""source"": {""url"": ""https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored"", ""name"": ""Open LLM Leaderboard""}}, {""task"": {""type"": ""text-generation"", ""name"": ""Text Generation""}, ""dataset"": {""name"": ""GSM8k (5-shot)"", ""type"": ""gsm8k"", ""config"": ""main"", ""split"": ""test"", ""args"": {""num_few_shot"": 5}}, ""metrics"": [{""type"": ""acc"", ""value"": 68.39, ""name"": ""accuracy"", ""verified"": false}], ""source"": {""url"": ""https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored"", ""name"": ""Open LLM Leaderboard""}}]}], ""config"": {""architectures"": [""LlamaForCausalLM""], ""model_type"": ""llama"", ""tokenizer_config"": {""bos_token"": ""<|begin_of_text|>"", ""chat_template"": ""{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}"", ""eos_token"": ""<|eot_id|>"", ""pad_token"": ""<|end_of_text|>""}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00001-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00002-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00003-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00004-of-00004.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [""KBaba7/Quant"", ""bhaskartripathi/LLM_Quantization"", ""totolook/Quant"", ""FallnAI/Quantize-HF-Models"", ""DevTzar/Orenguteng-Llama-3-8B-Lexi-Uncensored"", ""ruslanmv/convert_to_gguf"", ""SlyFox29/Quaken_Llama"", ""SlyFox29/Colonial_Llama"", ""AiJoker/Orenguteng-Llama-3-8B-Lexi-Uncensored"", ""basenban/Orenguteng-Llama-3-8B-Lexi-Uncensored"", ""Animus2018/Orenguteng-Llama-3-8B-Lexi-Uncensored"", ""tempwuefasdf/Orenguteng-Llama-3-8B-Lexi-Uncensored"", ""saumyadave24/hi"", ""asdaswadefswefr/sadbot2"", ""kaeyumir/Orenguteng-Llama-3-8B-Lexi-Uncensored"", ""abmSS/Orenguteng-Llama-3-8BKNK-Lexi-Uncensored"", ""Zadei/Orenguteng-Llama-3-8B-Lexi-Uncensored"", ""thara002/test"", ""UNION-KevinLan/Orenguteng-Llama-3-8B-Lexi-Uncensored"", ""nananie143/advanced-reasoning"", ""nananie143/agentic-system"", ""OscarFAI/inference"", ""K00B404/LLM_Quantization""], ""safetensors"": {""parameters"": {""F32"": 805306368, ""BF16"": 7224954880}, ""total"": 8030261248}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-05-27 06:16:40+00:00"", ""cardData"": ""license: llama3\ntags:\n- uncensored\n- llama3\n- instruct\n- open\nmodel-index:\n- name: Llama-3-8B-Lexi-Uncensored\n results:\n - task:\n type: text-generation\n name: Text Generation\n dataset:\n name: AI2 Reasoning Challenge (25-Shot)\n type: ai2_arc\n config: ARC-Challenge\n split: test\n args:\n num_few_shot: 25\n metrics:\n - type: acc_norm\n value: 59.56\n name: normalized accuracy\n verified: false\n source:\n url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored\n name: Open LLM Leaderboard\n - task:\n type: text-generation\n name: Text Generation\n dataset:\n name: HellaSwag (10-Shot)\n type: hellaswag\n split: validation\n args:\n num_few_shot: 10\n metrics:\n - type: acc_norm\n value: 77.88\n name: normalized accuracy\n verified: false\n source:\n url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored\n name: Open LLM Leaderboard\n - task:\n type: text-generation\n name: Text Generation\n dataset:\n name: MMLU (5-Shot)\n type: cais/mmlu\n config: all\n split: test\n args:\n num_few_shot: 5\n metrics:\n - type: acc\n value: 67.68\n name: accuracy\n verified: false\n source:\n url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored\n name: Open LLM Leaderboard\n - task:\n type: text-generation\n name: Text Generation\n dataset:\n name: TruthfulQA (0-shot)\n type: truthful_qa\n config: multiple_choice\n split: validation\n args:\n num_few_shot: 0\n metrics:\n - type: mc2\n value: 47.72\n verified: false\n source:\n url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored\n name: Open LLM Leaderboard\n - task:\n type: text-generation\n name: Text Generation\n dataset:\n name: Winogrande (5-shot)\n type: winogrande\n config: winogrande_xl\n split: validation\n args:\n num_few_shot: 5\n metrics:\n - type: acc\n value: 75.85\n name: accuracy\n verified: false\n source:\n url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored\n name: Open LLM Leaderboard\n - task:\n type: text-generation\n name: Text Generation\n dataset:\n name: GSM8k (5-shot)\n type: gsm8k\n config: main\n split: test\n args:\n num_few_shot: 5\n metrics:\n - type: acc\n value: 68.39\n name: accuracy\n verified: false\n source:\n url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored\n name: Open LLM Leaderboard"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""_id"": ""662824c0bfeb2ea792d69f33"", ""modelId"": ""Orenguteng/Llama-3-8B-Lexi-Uncensored"", ""usedStorage"": 17671168912}",0,"https://huggingface.co/PrunaAI/Orenguteng-Llama-3-8B-Lexi-Uncensored-HQQ-2bit-smashed, https://huggingface.co/scoliono/groupchat_lora_lexi_8b, https://huggingface.co/canTooDdev/LlamaWalter2",3,"https://huggingface.co/DesDea1/c101b3c6-dc54-441b-bc5c-d61fd0478fa9, https://huggingface.co/dada22231/5589168d-ec89-491d-b56d-03288e102c16, https://huggingface.co/nblinh63/5589168d-ec89-491d-b56d-03288e102c16, https://huggingface.co/eeeebbb2/add2f783-1808-4e94-9d46-4cfcf3b56992, https://huggingface.co/eeeebbb2/b2237d4d-aa9b-4276-ac4a-8a47f85a31e8, https://huggingface.co/laquythang/e4815ca4-4e52-4e25-96c3-9429db1ba0c6, https://huggingface.co/nblinh63/e4815ca4-4e52-4e25-96c3-9429db1ba0c6, https://huggingface.co/nblinh/e4815ca4-4e52-4e25-96c3-9429db1ba0c6, https://huggingface.co/eeeebbb2/e4815ca4-4e52-4e25-96c3-9429db1ba0c6, https://huggingface.co/dada22231/e4815ca4-4e52-4e25-96c3-9429db1ba0c6, https://huggingface.co/dada22231/8930a01c-082f-4c01-82c3-703887d7b3dc, https://huggingface.co/dada22231/e3c2746b-40eb-46bd-b52a-a864ae679571, https://huggingface.co/kooff11/e3c2746b-40eb-46bd-b52a-a864ae679571, https://huggingface.co/DeepDream2045/e15bb719-ea8f-46ea-8290-e5573063df0e, https://huggingface.co/vdos/e15bb719-ea8f-46ea-8290-e5573063df0e, https://huggingface.co/Rodo-Sami/e15bb719-ea8f-46ea-8290-e5573063df0e, https://huggingface.co/dada22231/e15bb719-ea8f-46ea-8290-e5573063df0e, https://huggingface.co/eeeebbb2/e15bb719-ea8f-46ea-8290-e5573063df0e, https://huggingface.co/Rodo-Sami/4f6d5dbe-c5bf-469a-a8bd-ac53a38104c1, https://huggingface.co/vdos/4f6d5dbe-c5bf-469a-a8bd-ac53a38104c1, https://huggingface.co/dimasik1987/5aa407fb-7336-4e21-87a7-80e987ee6437, https://huggingface.co/thalllsssss/5aa407fb-7336-4e21-87a7-80e987ee6437, https://huggingface.co/dimasik1987/74f5bf43-4a1b-44bb-9b95-6b5631ccfc3e, https://huggingface.co/lesso07/74f5bf43-4a1b-44bb-9b95-6b5631ccfc3e, https://huggingface.co/VERSIL91/a5a501fc-25a9-421c-a345-beb689b39e8a, https://huggingface.co/VERSIL91/9b3bfef1-1223-474e-97ff-c198ff0ab42c, https://huggingface.co/tarabukinivan/9b3bfef1-1223-474e-97ff-c198ff0ab42c, https://huggingface.co/tuanna08go/50bac6aa-85a2-403c-a522-11eb1be31abc, https://huggingface.co/VERSIL91/50bac6aa-85a2-403c-a522-11eb1be31abc, https://huggingface.co/lesso07/1e612fba-6f1d-4912-b2de-3e3388cd2164, https://huggingface.co/VERSIL91/1e612fba-6f1d-4912-b2de-3e3388cd2164, https://huggingface.co/nadejdatarabukina/d4c8a02e-2134-4039-b41e-e6d99462824d, https://huggingface.co/lesso08/d4c8a02e-2134-4039-b41e-e6d99462824d, https://huggingface.co/dimasik2987/d4c8a02e-2134-4039-b41e-e6d99462824d, https://huggingface.co/kokovova/15bf3ed5-8661-48f5-9d5d-a7047f62ad09, https://huggingface.co/lesso08/15bf3ed5-8661-48f5-9d5d-a7047f62ad09, https://huggingface.co/lesso02/15bf3ed5-8661-48f5-9d5d-a7047f62ad09, https://huggingface.co/hongngo/15bf3ed5-8661-48f5-9d5d-a7047f62ad09, https://huggingface.co/lesso05/d4c8a02e-2134-4039-b41e-e6d99462824d, https://huggingface.co/lesso07/15bf3ed5-8661-48f5-9d5d-a7047f62ad09, https://huggingface.co/lesso01/9eb6fe53-3d22-4f8a-aeac-2491baa2356c, https://huggingface.co/dzanbek/9eb6fe53-3d22-4f8a-aeac-2491baa2356c, https://huggingface.co/thaffggg/a0415c06-a436-4903-a853-cf2d53b0beff, https://huggingface.co/tarabukinivan/217ebb40-daa9-4e00-99de-eea1375e7a57, https://huggingface.co/lesso05/217ebb40-daa9-4e00-99de-eea1375e7a57, https://huggingface.co/lesso01/217ebb40-daa9-4e00-99de-eea1375e7a57, https://huggingface.co/hongngo/37f282ca-c51b-4c80-abbd-08ec5e67729e, https://huggingface.co/ivangrapher/ff34825d-d226-4b29-80c4-e1fcab3fbe85, https://huggingface.co/lesso02/ff34825d-d226-4b29-80c4-e1fcab3fbe85, https://huggingface.co/tuanna08go/ff34825d-d226-4b29-80c4-e1fcab3fbe85, https://huggingface.co/tarabukinivan/c5b2002e-a8a7-47f8-92d2-1145f7771181, https://huggingface.co/lesso06/c5b2002e-a8a7-47f8-92d2-1145f7771181, https://huggingface.co/cunghoctienganh/c5b2002e-a8a7-47f8-92d2-1145f7771181, https://huggingface.co/bbytxt/c5b2002e-a8a7-47f8-92d2-1145f7771181, https://huggingface.co/lesso05/fea6a7f6-5768-46b2-8073-ff96f567f2c9, https://huggingface.co/ivangrapher/fea6a7f6-5768-46b2-8073-ff96f567f2c9, https://huggingface.co/VERSIL91/fea6a7f6-5768-46b2-8073-ff96f567f2c9, https://huggingface.co/lesso08/fea6a7f6-5768-46b2-8073-ff96f567f2c9, https://huggingface.co/lesso03/19c02379-52ce-4259-8001-4cb1e57279c2, https://huggingface.co/bbytxt/19c02379-52ce-4259-8001-4cb1e57279c2, https://huggingface.co/nttx/19c02379-52ce-4259-8001-4cb1e57279c2, https://huggingface.co/cunghoctienganh/b702275d-23b5-4897-a306-7e984db67567, https://huggingface.co/VERSIL91/83856434-3640-462d-b03b-f9640982389e, https://huggingface.co/nttx/536ba553-9701-4fb3-9d86-95f6b2679893, https://huggingface.co/duyphu/b702275d-23b5-4897-a306-7e984db67567, https://huggingface.co/chauhoang/b702275d-23b5-4897-a306-7e984db67567, https://huggingface.co/kokovova/f5a371c6-b598-4f49-9aaa-c335bac9d3cb, https://huggingface.co/nhung02/f5a371c6-b598-4f49-9aaa-c335bac9d3cb, https://huggingface.co/thaffggg/f5a371c6-b598-4f49-9aaa-c335bac9d3cb, https://huggingface.co/phungkhaccuong/f5a371c6-b598-4f49-9aaa-c335bac9d3cb, https://huggingface.co/lesso06/8ccd446a-0d47-4313-a2bf-6fea19607f54, https://huggingface.co/nbninh/8ccd446a-0d47-4313-a2bf-6fea19607f54, https://huggingface.co/VERSIL91/8ccd446a-0d47-4313-a2bf-6fea19607f54, https://huggingface.co/dzanbek/ceef632e-f2fa-449b-a77c-22dff3b23ef5, https://huggingface.co/dzanbek/ef0e7220-dcbb-4819-a649-c74b01532a33, https://huggingface.co/lesso08/ef0e7220-dcbb-4819-a649-c74b01532a33, https://huggingface.co/VERSIL91/ef0e7220-dcbb-4819-a649-c74b01532a33, https://huggingface.co/bbytxt/ef0e7220-dcbb-4819-a649-c74b01532a33, https://huggingface.co/phungkhaccuong/8ccd446a-0d47-4313-a2bf-6fea19607f54, https://huggingface.co/nbninh/7fa985bf-6be0-40a0-8f01-e6cdd4b8d15c, https://huggingface.co/duyphu/d191d1e5-202f-29b8-f2c9-95bb6e479ff7, https://huggingface.co/lesso06/2d46aef7-a5de-41b7-89c5-f635c4818888, https://huggingface.co/lesso11/d3034b29-4944-4718-a3df-4ef89ea15152, https://huggingface.co/dimasik87/3bb55dcd-fe5a-4e50-b35d-bbe551edee20, https://huggingface.co/dzanbek/272fa23f-5941-44c4-9f54-bc964f647a9a, https://huggingface.co/chauhoang/8c1f3292-1794-6e6c-7890-503da1c075b9, https://huggingface.co/duyphu/0597210d-b1e7-fd1e-21e0-79eaab9ae9fb, https://huggingface.co/phungkhaccuong/7f7b6935-fd00-8b8e-f429-eb26353d3ffb, https://huggingface.co/cunghoctienganh/0185bc9c-7e35-40cb-aade-a081498f9cd7, https://huggingface.co/phungkhaccuong/eff65c60-daeb-4604-5781-a6c80e44f3ee, https://huggingface.co/filipesantoscv11/d6f48720-956e-44a7-ba9a-a69a061ada8d, https://huggingface.co/cunghoctienganh/1d015be7-0240-4846-a1a6-44987d920b8c, https://huggingface.co/lhong4759/d1c05990-8228-4946-bd35-5567ce1f996e, https://huggingface.co/filipesantoscv11/8bbe6f56-6033-4962-b7fa-45d1f75ed594, https://huggingface.co/lhong4759/29c9a7fb-e34c-4e02-bd97-bd07548e5853, https://huggingface.co/cunghoctienganh/804fc515-e956-4171-b539-ff2acb27a802, https://huggingface.co/VERSIL91/947b25e3-b276-4a08-8875-e5b98a03e2b8, https://huggingface.co/kk-aivio/decea65c-2daa-4bdd-b988-57df9b8f720e, https://huggingface.co/duyphu/c8477066-313e-4f43-fc45-e63527d89911, https://huggingface.co/filipesantoscv11/f6935419-ce8d-44e9-ac9d-0938bd53a60a, https://huggingface.co/phungkhaccuong/1da9e8d4-ae8a-bacd-8a79-38d9f0206cfa, https://huggingface.co/cunghoctienganh/a0882e49-30f6-4419-b070-58c89ca5994c, https://huggingface.co/nttx/9c8aa6ce-d6d8-48c0-8351-fd0d3dfc41ad, https://huggingface.co/thalllsssss/822a776d-c9bb-4850-a280-9cd752f236c4, https://huggingface.co/dzanbek/d30fe830-0e74-4679-87dd-4d6365915a46, https://huggingface.co/tuanna08go/ce8c1885-4011-1f8f-61d8-eb10d1daba72, https://huggingface.co/lesso03/b79decec-9467-4cb4-97b9-f12fe5d9856d, https://huggingface.co/lesso11/7083f0d9-6dc2-47c2-8cc7-b5a8f0e18268, https://huggingface.co/nhoxinh/b4d0121f-10ed-4727-b51a-733377367939, https://huggingface.co/lesso12/767d0f7c-1368-419a-b397-aa4a1a2b8141, https://huggingface.co/chauhoang/b7f88dd5-46c5-d6dd-7978-67587574c893, https://huggingface.co/prxy5605/2b79dd5a-8504-42db-9927-b0eec9004488, https://huggingface.co/prxy5604/974ea3a8-4175-4aff-a043-ac1e8915ed7d, https://huggingface.co/mrhunghd/ff823da0-8e2b-47f2-a114-4630f8ace140, https://huggingface.co/vermoney/aa62e62e-cd1d-45d0-9a23-cc1e5758254e, https://huggingface.co/nhung01/d5c74986-024c-48ab-8372-4af6df2a60dc, https://huggingface.co/chauhoang/351901c8-9a9c-4e58-83a3-39c60859993c, https://huggingface.co/VERSIL91/25e91404-fe01-4312-a741-3956a59f77f2, https://huggingface.co/FatCat87/taopanda-1_279e5cfb-d198-4bf0-8895-5af873459233, https://huggingface.co/nttx/72879849-db2b-49dd-ae28-983818de72e1, https://huggingface.co/prxy5607/58e13912-36fe-41e6-b320-ed0b12501097, https://huggingface.co/kokovova/27c36b34-c78d-4218-8a6d-4b9f8b6571fa, https://huggingface.co/phungkhaccuong/9842c1d0-6a0c-4e12-8f03-406af2ec0a3e, https://huggingface.co/aleegis10/774384b1-2027-4c99-8d1e-5c7420c09af9, https://huggingface.co/cunghoctienganh/17de06bc-8dc5-42a2-8e5d-9b8af6efecb8, https://huggingface.co/nblinh/87219660-6063-4a2d-923a-067ccba8718d, https://huggingface.co/hongngo/6428d252-a689-42c7-9288-19190efe172f, https://huggingface.co/thakkkkkk/1472366c-1474-4e23-9b6f-e917cacaed31, https://huggingface.co/error577/634fe6e7-ba15-40a0-84cd-c93ce43b7688, https://huggingface.co/0x1202/b1a15532-d25e-44df-b90a-8083ef674b2f, https://huggingface.co/adammandic87/59a69d45-742f-462f-8f72-f0c09ff0fc21, https://huggingface.co/prxy5608/363d02b1-884e-48cc-9864-b4be6c07aa89, https://huggingface.co/VERSIL91/cd41e37a-3f7e-46a6-8ba7-d8b4f2155726, https://huggingface.co/marialvsantiago/50df958b-6c6c-4bbb-96c1-cdfdb3983a97, https://huggingface.co/nblinh63/f6afc9f5-940e-4854-802f-2f55b6c0dfe0, https://huggingface.co/dzanbek/b955aef4-8949-4ef3-b748-e55af8713ba6, https://huggingface.co/lesso14/3bd2c9ad-cbe9-45aa-b3ba-a35707cfb659, https://huggingface.co/adammandic87/ab7997db-c4af-4085-ad6e-afc46f65f1dd, https://huggingface.co/chauhoang/fe6f0dae-5342-4bc8-8a1f-f80d21845e48, https://huggingface.co/kokovova/dfcef1f5-9900-4e8f-9e5c-b2d3a0907a5c, https://huggingface.co/mrHunghddddd/b51de4ec-4808-43cb-93c3-0f3cd9803d7e, https://huggingface.co/dimasik87/2bcb1699-a32d-4c49-abd3-0028f372c4e8, https://huggingface.co/dimasik1987/3d767089-d3f6-424e-902f-1e490617d192, https://huggingface.co/nhoxinh/7dbd9515-0d99-48d1-8ad7-f9ea9d7e6e01, https://huggingface.co/Paladiso/e307da6f-722d-4008-b2b6-3ae184f063ee, https://huggingface.co/prxy5608/c4767a78-790f-4877-a512-a320a2b24062, https://huggingface.co/kokovova/deee9727-f8a7-4ce0-bfc8-a025fb41db6a, https://huggingface.co/filipesantoscv11/d98c90de-4655-41b2-9d06-681e68b98db0, https://huggingface.co/lhong4759/eba0be74-d070-4fd7-8dff-b56e64b4cab2, https://huggingface.co/chauhoang/b35542fa-eda8-459c-8040-ff872c89c330, https://huggingface.co/thalllsssss/479c7186-b280-4ef5-a30f-419e77ddcd70, https://huggingface.co/adammandic87/a7526096-751d-4831-97c7-ddd04420860d, https://huggingface.co/adammandic87/392372e1-a883-48a4-a5fd-1228479e4ee8, https://huggingface.co/0x1202/b79042e9-c4aa-4947-af63-001439792f03, https://huggingface.co/thalllsssss/f12b6437-b14c-4d68-aeeb-a3fffed33204, https://huggingface.co/lesso05/1f73b57d-7d3e-42d7-91a6-716c0df80d29, https://huggingface.co/Best000/2b443fa3-7ee8-4e00-82b6-b279668fa2e5, https://huggingface.co/tuanna08go/cafd5aed-0704-4eaa-b6c3-1e52d308f4f6, https://huggingface.co/vermoney/dcbebbf3-5e74-4303-88ac-2e90ff4350ed, https://huggingface.co/lesso07/2bc6b804-5e10-43af-93b7-6119619d4865, https://huggingface.co/dimasik87/09eb108d-4d2c-41e4-ac34-f57a5dadfc9c, https://huggingface.co/thangla01/d6c82f6f-38d7-4dfe-88cc-47934c8d003b, https://huggingface.co/dzanbek/bc0e367c-0592-4b01-a54f-4fd9af043723, https://huggingface.co/lesso14/2eafcc2b-34bf-461d-ad55-78d59697b25e, https://huggingface.co/havinash-ai/823bdeeb-5d0b-4f18-8ebc-ac9cffa9fcc4, https://huggingface.co/great0001/0e055e0f-abf8-40e0-baf7-2cefd5c93eb3, https://huggingface.co/kk-aivio/36fa3a7f-931d-4a88-9c30-07ed2865b2a0, https://huggingface.co/tarabukinivan/ee964427-1cfe-4271-97bd-4d8150d53a35, https://huggingface.co/mrHunghddddd/f5ecffc9-b7d5-462d-9d6e-7d19f5cee13c, https://huggingface.co/thangla01/e887bd4b-58c1-4c1f-9bb8-c5eaeab76359, https://huggingface.co/dzanbek/39717d07-1996-41ab-a2bb-30c9048abb79, https://huggingface.co/lesso01/1b4f4a4d-62c7-455a-afc8-62ed8dda02f9, https://huggingface.co/lesso13/1e92a533-fe72-4dea-a29e-1b84ce3c3942, https://huggingface.co/great0001/f2642680-9820-4049-b96d-0182e0ebacf5, https://huggingface.co/chauhoang/2596f300-21c5-4b74-ba07-ce69ac6c8bb4, https://huggingface.co/duyphu/36dc4db4-5a76-4ea9-9e15-42ff61911844, https://huggingface.co/prxy5605/098d1a5a-5801-4a95-bb78-baf302eb8ad7, https://huggingface.co/lesso10/5442a177-d921-4d91-9c03-6eedfd535043, https://huggingface.co/nhung01/a3ab2628-3072-4382-a0e0-06bd37adc5ed, https://huggingface.co/nttx/b0145126-1b95-4959-9715-c6a828f48822, https://huggingface.co/duyphu/7b7d01be-2a9a-4da2-95dd-b16cbb975fc2, https://huggingface.co/lesso17/448acd22-fe99-4277-b378-b66066353512, https://huggingface.co/thaffggg/4aaa1c5c-8bef-4b75-98dc-585f050a605b, https://huggingface.co/nttx/24e206ea-db0c-4b0f-8114-b1408e712349, https://huggingface.co/nhungphammmmm/9cdf2d91-4b98-4839-99ea-55b799b02ae2, https://huggingface.co/hongngo/14572a1f-422e-4ad1-b681-1a7a37eac195, https://huggingface.co/lhong4759/fa7f924e-5e9a-4b7a-9cf1-36a037cfd92c, https://huggingface.co/lesso04/f681eb1b-9035-4097-b398-36b35fd433ea, https://huggingface.co/ClarenceDan/20665300-5c57-40b7-afb0-ae0a8629d55b, https://huggingface.co/lesso17/0cc28810-bb73-4af6-80aa-c1faea907edc, https://huggingface.co/cunghoctienganh/eabee629-5417-40d1-b350-b688cc9b395b, https://huggingface.co/minhnguyennnnnn/6cc5490d-760b-41db-8c55-3ba405f5bb5f, https://huggingface.co/nblinh/a068eee1-f185-4300-b139-7db918e74bdb, https://huggingface.co/tryingpro/fff4bb41-fbd1-4606-80cd-29d5047ddf58, https://huggingface.co/hongngo/99d56d7e-c912-450c-9c7e-1673741da56d, https://huggingface.co/mrhunghd/b46793f9-6b2c-4bce-a727-ccee5b2ed0ba, https://huggingface.co/nhung03/d0aeafa1-92ab-45ba-8279-d91d79083014, https://huggingface.co/dimasik87/f6cb407e-779d-4ac8-948d-936ca7ae0140, https://huggingface.co/lesso14/8eb64cf5-9343-4cfe-a62d-6f21b5fcc610, https://huggingface.co/duyphu/f04b6d2e-d964-463f-8e64-56594e30b835, https://huggingface.co/ClarenceDan/02169017-6ccb-4b8a-aed4-045c6cbea35f, https://huggingface.co/prxy5604/9a79ad87-143c-4112-826d-17de569acc7b, https://huggingface.co/ClarenceDan/8a9e61e6-77cc-43cc-8b7f-5d6bd3a47368, https://huggingface.co/duyphu/dd99cfae-4152-4304-b09a-31a00951788f, https://huggingface.co/bbytxt/b75078cf-3684-441e-bd0e-6229f5588377, https://huggingface.co/eddysang/4ebfb258-edc4-4a06-9f79-9cc8fd87c956, https://huggingface.co/aleegis11/afe4bad5-274c-4673-9e9c-09057cffabae, https://huggingface.co/prxy5606/99a88bee-0683-462c-9ba0-36b004cece89, https://huggingface.co/aleegis09/ea36b8db-3107-46e2-9710-8040b5b07b59, https://huggingface.co/prxy5608/7dac049d-ecb7-4466-bae7-c8db9bb455f2, https://huggingface.co/prxy5606/18f04b94-dd45-4ae2-ae06-ecae560f913f, https://huggingface.co/prxy5608/d7edf5a8-da9f-49c4-abaf-477279cdcc80, https://huggingface.co/aleegis10/7e903844-4623-4593-a49c-f5f63be9eb3c, https://huggingface.co/denbeo/1f2e7ce1-66d0-47cd-87a6-79a86e9cd603, https://huggingface.co/mrHunghddddd/7735e12e-8e0c-4441-b324-1b721387f8d4, https://huggingface.co/kk-aivio/98f9cbf5-961f-4aca-80f3-c5555d8bcd9b, https://huggingface.co/nhung02/e70346f3-df2e-4c9e-b0ee-77b0c0dd4a53, https://huggingface.co/lesso03/539f11df-2469-4a67-9924-40fb548c4bb9, https://huggingface.co/adammandic87/7b35d4dd-d862-4984-b66e-42a4e11d404b, https://huggingface.co/JacksonBrune/0e987a56-97ef-4ebb-993c-62a8c77a762d, https://huggingface.co/adammandic87/014d6916-9095-4085-aaa2-2a9e63da7f61, https://huggingface.co/chauhoang/0767ca9f-5a86-4f5e-9597-a16f678da1be, https://huggingface.co/tryingpro/34fedf70-94ce-42f7-836f-c3ee8710c95a, https://huggingface.co/nghiatrannnnnn/512a114c-6d40-44f8-a819-61e951e26bb3, https://huggingface.co/nhunglaaaaaaa/8bda2c6e-cca4-4d6c-bdc8-0784355e392c, https://huggingface.co/lesso09/0fc82aab-59d1-48c9-bb60-f04c04fac4e5, https://huggingface.co/lesso/53ffda63-737e-45a1-a819-f90080283be2, https://huggingface.co/demohong/69ef178a-2bf3-4f2e-8cf2-ed5cf25ba18b, https://huggingface.co/mrHunghddddd/79da53b3-2277-410c-a08d-0441604c1d25, https://huggingface.co/thalllsssss/fe5ca7ec-36ef-42d8-b2d4-72877bc4597c, https://huggingface.co/nhung03/9836cde7-5566-4ef9-9d05-28d754fd1fa5, https://huggingface.co/nhung01/ce601c6d-f85d-4322-900a-4843d391639e, https://huggingface.co/tuantmdev/84359d7e-b509-4ff9-8bed-0d22fd92b1ca, https://huggingface.co/nbninh/4fc177ad-16a7-46ff-9c51-303e9338bda8, https://huggingface.co/cunghoctienganh/a1bef39b-65cf-4d6b-991a-da567c6c8c41, https://huggingface.co/lhong4759/26d24d5d-48d4-4f79-b3ec-bd100ad807ed, https://huggingface.co/lesso12/193b0bd7-e478-452d-b5cd-ecce9fec986b, https://huggingface.co/daniel40/aaab34a9-3ba4-4596-8cce-5eee193fd508, https://huggingface.co/aleegis10/05a07b91-53ab-454c-8c8a-2e82fb43592d, https://huggingface.co/aleegis11/b2fef5fc-aec3-418b-b721-5ab6ee0bb573, https://huggingface.co/bbytxt/20beb952-12e3-4b9b-a1f3-1a3f94d8e798, https://huggingface.co/trangtrannnnn/a77410da-cb3b-40eb-a485-f50d4ef0b664, https://huggingface.co/datlaaaaaaa/bf99c9d0-5c7c-4534-bd64-48d748cc71ef, https://huggingface.co/thangla01/fca7c77b-f48b-40b1-8d6c-361b3476b051, https://huggingface.co/demohong/be5bf364-34a9-4792-b938-7ae58cc17e24, https://huggingface.co/myhaaaaaaa/341644b7-b9fd-4fe9-8873-02292876215b, https://huggingface.co/nhung02/eaf5a95f-15be-46bf-8aa4-5a24b02a3476, https://huggingface.co/aseratus1/4989f67b-c847-4177-af12-df137cf633bc, https://huggingface.co/nttx/7d59fed9-8564-4996-874f-f5884135c1b4, https://huggingface.co/daniel40/e42e4607-e2d3-418b-be24-a89ed2a70d60, https://huggingface.co/nghiatrannnnnn/7456c8f0-92e4-431d-b9c1-bb06db8067b1, https://huggingface.co/nghiatrannnnnn/030f1c7f-ff4e-4cf6-b23d-01d69b545282, https://huggingface.co/nhungphammmmm/06524e78-c8b3-4960-a912-92de39d451d1, https://huggingface.co/Romain-XV/c1c9a393-ad22-468d-a98e-1dcb95b2d0ec, https://huggingface.co/nhung03/922f138a-b3e6-4a86-8494-91120aa3d847, https://huggingface.co/lesso07/d171a9ac-bdba-45c5-b8c9-01c274f5cd32, https://huggingface.co/tarabukinivan/dab39e16-16dc-416c-9022-3ad4b4d9d8a7, https://huggingface.co/nhoxinh/f4e918c4-32e6-4dbc-a546-6cabe8f6ba27, https://huggingface.co/abaddon182/eac504e2-99ab-4d7a-a27a-b86e42c79af8, https://huggingface.co/adammandic87/032cbbb5-bf71-40df-bff5-a79fb44167c5, https://huggingface.co/lesso06/dcffe245-421b-4d47-8057-fa36f3805411, https://huggingface.co/0x1202/c3cef6b8-12c1-48bf-8d86-335a840ec84a, https://huggingface.co/minhnguyennnnnn/0a982b54-d90e-4a3c-9e7f-11a4dedac0c1, https://huggingface.co/cunghoctienganh/90e633df-e0f4-4ba7-b984-9015145b9db5, https://huggingface.co/minhtrannnn/1f47c03d-1215-4271-8dbd-86d80dbc11d8, https://huggingface.co/nhung01/27d2e254-e373-4613-85f6-d3c9e86fd5ca, https://huggingface.co/laquythang/74e9dd27-7cb4-4910-a8bf-13b76ad02f5e, https://huggingface.co/thaffggg/3a5d88b3-db9c-461c-b396-76b88a6cee39, https://huggingface.co/laquythang/eeafba04-c9ef-4dbc-921c-7d14d109a317, https://huggingface.co/lesso/9af11431-ae6c-40ba-8a7c-ecf988668da1, https://huggingface.co/ancient41/31fd523e-8b03-4b09-a487-0e568879568f, https://huggingface.co/mamung/3f49b8e9-e2b8-4a3b-b363-8483b5368549, https://huggingface.co/nhung03/bd08913d-aff9-4e52-9536-39d293fd521b, https://huggingface.co/laquythang/1122a8d5-25bd-4613-9a59-16350b1cf8fa, https://huggingface.co/robiulawaldev/ddd02f7c-9454-4722-b751-841b484a71a4, https://huggingface.co/adammandic87/491009d8-1596-49c7-b9f6-26bf8ab5a711, https://huggingface.co/kk-aivio/cd52ae9a-9eb7-4cfe-bdad-e281fa438605, https://huggingface.co/cilooor/a975dc64-d122-46f1-8755-130013566f75, https://huggingface.co/ancient41/d43662eb-c176-493f-a566-011f46f3f100, https://huggingface.co/prxy5604/1d0b0ed5-2d83-4cdf-8ecd-1e342748bada, https://huggingface.co/aleegis12/6d7ba68c-b4dd-4516-b48e-7acbcb8b242a, https://huggingface.co/alchemist69/fb3c878d-6daa-474c-86e9-521c77cfadc6, https://huggingface.co/lesso/10ccc4f9-e48e-4823-a9b8-d8686863fefe, https://huggingface.co/lesso/b40a4d62-ce92-4502-a0c6-0abe2c9e48a4, https://huggingface.co/arcwarden46/66c7820b-9388-468a-8095-63317a4ed8e8, https://huggingface.co/aleegis12/1192696c-aa12-4e64-b0fe-dcccc7dd37e3, https://huggingface.co/alchemist69/50e42555-c03c-4c8c-8f49-ecc602637af1, https://huggingface.co/bane5631/a649fb20-63c4-4576-88b2-d4c959b78475, https://huggingface.co/laquythang/3c244461-d68f-489b-bc6d-ece553ad708c, https://huggingface.co/lesso/244bde5e-38f7-4513-b43d-638ced78af2d, https://huggingface.co/eddysang/30d14f86-1bea-48d7-97b1-2e4c3d700e53, https://huggingface.co/arcwarden46/104eb290-e68b-4823-89e0-65029759e501, https://huggingface.co/eageringdev/945c81e4-eeb4-43e3-bce2-0f91952c347e, https://huggingface.co/cimol/92560246-778b-464e-a130-85faabcd455c, https://huggingface.co/lesso04/4d8a83c6-b15c-4324-9571-2f06a1409031, https://huggingface.co/lesso17/0c47821e-1e59-4629-8b79-116c2fbf737a, https://huggingface.co/lesso06/e2cfacf6-0034-4587-8961-098036d0f7fc, https://huggingface.co/romainnn/d8abd64a-a056-4dc3-acae-1063719be070, https://huggingface.co/nttx/b1020384-4fed-43f0-b098-e7f1d056ec24, https://huggingface.co/lesso07/e1c6f8e8-5823-422c-ae61-dbb86493c306, https://huggingface.co/lesso12/f10b8915-f0a3-4d58-bfc9-e9ab4e7cae90, https://huggingface.co/lesso02/234de95a-352d-4a70-b496-8cbc55e31388, https://huggingface.co/lesso06/41b2b81b-40ab-4e64-bec1-5826d67fbb58, https://huggingface.co/lesso01/addbdac6-1161-45e6-8f1a-9e89695db7dd, https://huggingface.co/lesso10/94fa69ad-7f10-4114-aea9-549d491b720a, https://huggingface.co/lesso18/7155a30e-2eae-48cf-abbd-d38ed410e9ff, https://huggingface.co/lesso14/eb6d5aee-dd12-4a7f-af40-054232eb7660, https://huggingface.co/lesso09/3cdac240-e0c9-4569-8046-139a0fa1d0b7, https://huggingface.co/lesso03/c25a6821-70d3-41d9-b043-200034ef0df0, https://huggingface.co/lesso02/1d0e5272-383b-476f-8ea2-1767043388d1, https://huggingface.co/lesso09/8ba6a94d-6a12-48f3-9922-ebba4fe79628, https://huggingface.co/lesso18/975b7468-c0cf-46fd-b9ef-480837bfe221, https://huggingface.co/lesso13/5af29646-c4ef-4955-b0d7-801598948193, https://huggingface.co/lesso02/1e748813-92c4-4308-9bef-4debf187f3da, https://huggingface.co/lesso13/1c45cc1c-13be-45d5-a068-1fcfa1d24bbc, https://huggingface.co/lesso09/bc6b0b68-cc10-4af2-90ab-0a3ea4118781, https://huggingface.co/lesso02/48ce706e-9aac-41aa-b28b-e21b7660868f, https://huggingface.co/lesso05/0befb973-0bc9-4f06-ae5f-ab32f5900322, https://huggingface.co/lesso03/30d96abc-3305-4f66-8d3d-3a95a991e512, https://huggingface.co/Alphatao/6da1faf7-1428-4419-b21c-b21eec14b3ea, https://huggingface.co/lesso05/1f25bc1f-d773-4387-9e3b-fa35936c5193, https://huggingface.co/lesso02/db7fa05d-a82f-43dc-9084-3d41b53e204b, https://huggingface.co/Alphatao/aff88e51-b725-41ad-bf68-2c6c7bac164f, https://huggingface.co/lesso13/b18a3101-7e5c-4e6b-ba1f-ca8aa0c9c28f, https://huggingface.co/lesso12/4c0ec110-d24c-48d6-827a-afeaf1b7014a, https://huggingface.co/Alphatao/90451222-f43c-4606-84e4-bb92cc585d7b, https://huggingface.co/Alphatao/e5e9e1db-edd7-4baf-b538-728f32f0be9a, https://huggingface.co/Alphatao/d96d18be-1c39-49c1-b429-f6623f4ab502, https://huggingface.co/lesso05/385d9043-228f-412c-be31-6b58cec7eeab, https://huggingface.co/lesso01/721b0e6f-4a4b-4bf6-b14b-8ae83ae43adb, https://huggingface.co/Alphatao/03f640f1-7d7d-40be-98fe-9441de6cce1a, https://huggingface.co/lesso05/a299b07e-5e51-477d-98d1-4b78090f3f7a, https://huggingface.co/lesso10/c4989afc-e02d-4ee8-ac14-abe295d2eec5, https://huggingface.co/lesso16/b8ef9103-bbf1-4f0e-aa78-9083b02a8b8c, https://huggingface.co/lesso07/cbba3e68-5c04-43e1-85e2-0d68078ce4d3, https://huggingface.co/sergioalves/d0f62694-a9a9-41ee-92fa-739328b8e778, https://huggingface.co/kokovova/30227d13-f5a0-44bd-980b-4a818385f65b, https://huggingface.co/vermoney/68063ba7-c5c1-461b-b3bf-97df947cfe3c, https://huggingface.co/annemiekebickleyoy/51cd8f4f-b436-4738-8e35-439e2016ba0c",339,"https://huggingface.co/QuantFactory/Llama-3-8B-Lexi-Uncensored-GGUF, https://huggingface.co/stephenlzc/Llama-3-8B-Lexi-Uncensored-Q2_K-GGUF, https://huggingface.co/mradermacher/Llama-3-8B-Lexi-Uncensored-GGUF, https://huggingface.co/solidrust/Llama-3-8B-Lexi-Uncensored-AWQ, https://huggingface.co/PrunaAI/Orenguteng-Llama-3-8B-Lexi-Uncensored-bnb-4bit-smashed, https://huggingface.co/Ayyystin/Llama-3-8B-Lexi-Uncensored-Q4_0-GGUF, https://huggingface.co/wyan/Llama-3-8B-Lexi-Uncensored-Q4_K_M-GGUF, https://huggingface.co/wyan/Llama-3-8B-Lexi-Uncensored-Q8_0-GGUF, https://huggingface.co/jdcrutchley/Llama-3-8B-Lexi-Uncensored-Q4_K_M-GGUF, https://huggingface.co/MaziyarPanahi/Llama-3-8B-Lexi-Uncensored-GGUF, https://huggingface.co/tensorblock/Llama-3-8B-Lexi-Uncensored-GGUF, https://huggingface.co/mradermacher/Lexi-Llama-3-8B-Uncensored-GGUF, https://huggingface.co/mradermacher/Lexi-Llama-3-8B-Uncensored-i1-GGUF, https://huggingface.co/wolflycanorcant/Llama-3-8B-Lexi-Uncensored-Q4_K_M-GGUF, https://huggingface.co/itlwas/Llama-3-8B-Lexi-Uncensored-Q4_K_M-GGUF, https://huggingface.co/mradermacher/Llama-3-8B-Lexi-Uncensored-i1-GGUF",16,"https://huggingface.co/theprint/Llama-3-8B-Lexi-Smaug-Uncensored, https://huggingface.co/mergekit-community/uncensored-mix, https://huggingface.co/Nhoodie/Meta-Llama-3-8b-Lexi-Uninstruct-function-calling-json-mode-Task-Arithmetic-v0.1, https://huggingface.co/Nhoodie/Meta-Llama-3-8b-Lexi-Uninstruct-function-calling-json-mode-Task-Arithmetic-v0.2A, https://huggingface.co/Nhoodie/Meta-Llama-3-8b-Extended-Lexi-Uninstruct-function-calling-json-mode-Task-Arithmetic-v0.0A, https://huggingface.co/Nhoodie/Meta-Llama-3-8b-Configurable-Lexi-Uninstruct-function-calling-json-mode-Task-Arithmetic-v0.0A, https://huggingface.co/Fischerboot/LLama3-Lexi-Aura-3Some-SLERP-SLERP, https://huggingface.co/Rupesh2/Llama-3.1-Uncensored-New, https://huggingface.co/Casual-Autopsy/L3-Luna-8B, https://huggingface.co/QuantFactory/L3-Luna-8B-GGUF, https://huggingface.co/ZeroXClem/Llama-3-Aetheric-Hermes-Lexi-Smaug-8B, https://huggingface.co/mav23/Llama-3-Aetheric-Hermes-Lexi-Smaug-8B-GGUF",12,"DevTzar/Orenguteng-Llama-3-8B-Lexi-Uncensored, FallnAI/Quantize-HF-Models, HuggingFaceH4/open_llm_leaderboard?query=Orenguteng/Llama-3-8B-Lexi-Uncensored, K00B404/LLM_Quantization, KBaba7/Quant, OscarFAI/inference, SlyFox29/Quaken_Llama, UNION-KevinLan/Orenguteng-Llama-3-8B-Lexi-Uncensored, Zadei/Orenguteng-Llama-3-8B-Lexi-Uncensored, abmSS/Orenguteng-Llama-3-8BKNK-Lexi-Uncensored, bhaskartripathi/LLM_Quantization, huggingface/InferenceSupport/discussions/82, ruslanmv/convert_to_gguf, totolook/Quant",14
138
+ PrunaAI/Orenguteng-Llama-3-8B-Lexi-Uncensored-HQQ-2bit-smashed,"---
139
+ thumbnail: ""https://assets-global.website-files.com/646b351987a8d8ce158d1940/64ec9e96b4334c0e1ac41504_Logo%20with%20white%20text.svg""
140
+ base_model: Orenguteng/Llama-3-8B-Lexi-Uncensored
141
+ metrics:
142
+ - memory_disk
143
+ - memory_inference
144
+ - inference_latency
145
+ - inference_throughput
146
+ - inference_CO2_emissions
147
+ - inference_energy_consumption
148
+ tags:
149
+ - pruna-ai
150
+ ---
151
+ <!-- header start -->
152
+ <!-- 200823 -->
153
+ <div style=""width: auto; margin-left: auto; margin-right: auto"">
154
+ <a href=""https://www.pruna.ai/"" target=""_blank"" rel=""noopener noreferrer"">
155
+ <img src=""https://i.imgur.com/eDAlcgk.png"" alt=""PrunaAI"" style=""width: 100%; min-width: 400px; display: block; margin: auto;"">
156
+ </a>
157
+ </div>
158
+ <!-- header end -->
159
+
160
+ [![Twitter](https://img.shields.io/twitter/follow/PrunaAI?style=social)](https://twitter.com/PrunaAI)
161
+ [![GitHub](https://img.shields.io/github/followers/PrunaAI?label=Follow%20%40PrunaAI&style=social)](https://github.com/PrunaAI)
162
+ [![LinkedIn](https://img.shields.io/badge/LinkedIn-Connect-blue)](https://www.linkedin.com/company/93832878/admin/feed/posts/?feedType=following)
163
+ [![Discord](https://img.shields.io/badge/Discord-Join%20Us-blue?style=social&logo=discord)](https://discord.gg/rskEr4BZJx)
164
+
165
+ # Simply make AI models cheaper, smaller, faster, and greener!
166
+
167
+ - Give a thumbs up if you like this model!
168
+ - Contact us and tell us which model to compress next [here](https://www.pruna.ai/contact).
169
+ - Request access to easily compress your *own* AI models [here](https://z0halsaff74.typeform.com/pruna-access?typeform-source=www.pruna.ai).
170
+ - Read the documentations to know more [here](https://pruna-ai-pruna.readthedocs-hosted.com/en/latest/)
171
+ - Join Pruna AI community on Discord [here](https://discord.gg/rskEr4BZJx) to share feedback/suggestions or get help.
172
+
173
+ ## Results
174
+
175
+ ![image info](./plots.png)
176
+
177
+ **Frequently Asked Questions**
178
+ - ***How does the compression work?*** The model is compressed with hqq.
179
+ - ***How does the model quality change?*** The quality of the model output might vary compared to the base model.
180
+ - ***How is the model efficiency evaluated?*** These results were obtained on NVIDIA A100-PCIE-40GB with configuration described in `model/smash_config.json` and are obtained after a hardware warmup. The smashed model is directly compared to the original base model. Efficiency results may vary in other settings (e.g. other hardware, image size, batch size, ...). We recommend to directly run them in the use-case conditions to know if the smashed model can benefit you.
181
+ - ***What is the model format?*** We use safetensors.
182
+ - ***What calibration data has been used?*** If needed by the compression method, we used WikiText as the calibration data.
183
+ - ***What is the naming convention for Pruna Huggingface models?*** We take the original model name and append ""turbo"", ""tiny"", or ""green"" if the smashed model has a measured inference speed, inference memory, or inference energy consumption which is less than 90% of the original base model.
184
+ - ***How to compress my own models?*** You can request premium access to more compression methods and tech support for your specific use-cases [here](https://z0halsaff74.typeform.com/pruna-access?typeform-source=www.pruna.ai).
185
+ - ***What are ""first"" metrics?*** Results mentioning ""first"" are obtained after the first run of the model. The first run might take more memory or be slower than the subsequent runs due cuda overheads.
186
+ - ***What are ""Sync"" and ""Async"" metrics?*** ""Sync"" metrics are obtained by syncing all GPU processes and stop measurement when all of them are executed. ""Async"" metrics are obtained without syncing all GPU processes and stop when the model output can be used by the CPU. We provide both metrics since both could be relevant depending on the use-case. We recommend to test the efficiency gains directly in your use-cases.
187
+
188
+ ## Setup
189
+
190
+ You can run the smashed model with these steps:
191
+
192
+ 0. Check requirements from the original repo Orenguteng/Llama-3-8B-Lexi-Uncensored installed. In particular, check python, cuda, and transformers versions.
193
+ 1. Make sure that you have installed quantization related packages.
194
+ ```bash
195
+ pip install hqq
196
+ ```
197
+ 2. Load & run the model.
198
+ ```python
199
+ from transformers import AutoModelForCausalLM, AutoTokenizer
200
+ from hqq.engine.hf import HQQModelForCausalLM
201
+ from hqq.models.hf.base import AutoHQQHFModel
202
+
203
+ try:
204
+ model = HQQModelForCausalLM.from_quantized(""PrunaAI/Orenguteng-Llama-3-8B-Lexi-Uncensored-HQQ-2bit-smashed"", device_map='auto')
205
+ except:
206
+ model = AutoHQQHFModel.from_quantized(""PrunaAI/Orenguteng-Llama-3-8B-Lexi-Uncensored-HQQ-2bit-smashed"")
207
+ tokenizer = AutoTokenizer.from_pretrained(""Orenguteng/Llama-3-8B-Lexi-Uncensored"")
208
+
209
+ input_ids = tokenizer(""What is the color of prunes?,"", return_tensors='pt').to(model.device)[""input_ids""]
210
+
211
+ outputs = model.generate(input_ids, max_new_tokens=216)
212
+ tokenizer.decode(outputs[0])
213
+ ```
214
+
215
+ ## Configurations
216
+
217
+ The configuration info are in `smash_config.json`.
218
+
219
+ ## Credits & License
220
+
221
+ The license of the smashed model follows the license of the original model. Please check the license of the original model Orenguteng/Llama-3-8B-Lexi-Uncensored before using this model which provided the base model. The license of the `pruna-engine` is [here](https://pypi.org/project/pruna-engine/) on Pypi.
222
+
223
+ ## Want to compress other models?
224
+
225
+ - Contact us and tell us which model to compress next [here](https://www.pruna.ai/contact).
226
+ - Request access to easily compress your own AI models [here](https://z0halsaff74.typeform.com/pruna-access?typeform-source=www.pruna.ai).","{""id"": ""PrunaAI/Orenguteng-Llama-3-8B-Lexi-Uncensored-HQQ-2bit-smashed"", ""author"": ""PrunaAI"", ""sha"": ""1dab39f5ed6b5c5e5b327c2c052198388c653594"", ""last_modified"": ""2024-08-02 15:56:58+00:00"", ""created_at"": ""2024-04-29 15:39:09+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 5, ""downloads_all_time"": null, ""likes"": 1, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""llama"", ""text-generation"", ""pruna-ai"", ""conversational"", ""base_model:Orenguteng/Llama-3-8B-Lexi-Uncensored"", ""base_model:finetune:Orenguteng/Llama-3-8B-Lexi-Uncensored"", ""autotrain_compatible"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: Orenguteng/Llama-3-8B-Lexi-Uncensored\nmetrics:\n- memory_disk\n- memory_inference\n- inference_latency\n- inference_throughput\n- inference_CO2_emissions\n- inference_energy_consumption\ntags:\n- pruna-ai\nthumbnail: https://assets-global.website-files.com/646b351987a8d8ce158d1940/64ec9e96b4334c0e1ac41504_Logo%20with%20white%20text.svg"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": null, ""config"": {""architectures"": [""LlamaForCausalLM""], ""model_type"": ""llama"", ""tokenizer_config"": {""bos_token"": ""<|begin_of_text|>"", ""chat_template"": ""{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}"", ""eos_token"": ""<|eot_id|>"", ""pad_token"": ""<|end_of_text|>""}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='plots.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='qmodel.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='results.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='smash_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-08-02 15:56:58+00:00"", ""cardData"": ""base_model: Orenguteng/Llama-3-8B-Lexi-Uncensored\nmetrics:\n- memory_disk\n- memory_inference\n- inference_latency\n- inference_throughput\n- inference_CO2_emissions\n- inference_energy_consumption\ntags:\n- pruna-ai\nthumbnail: https://assets-global.website-files.com/646b351987a8d8ce158d1940/64ec9e96b4334c0e1ac41504_Logo%20with%20white%20text.svg"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""_id"": ""662fbf1d14e87951eadab03c"", ""modelId"": ""PrunaAI/Orenguteng-Llama-3-8B-Lexi-Uncensored-HQQ-2bit-smashed"", ""usedStorage"": 4010664640}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=PrunaAI/Orenguteng-Llama-3-8B-Lexi-Uncensored-HQQ-2bit-smashed&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BPrunaAI%2FOrenguteng-Llama-3-8B-Lexi-Uncensored-HQQ-2bit-smashed%5D(%2FPrunaAI%2FOrenguteng-Llama-3-8B-Lexi-Uncensored-HQQ-2bit-smashed)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
227
+ scoliono/groupchat_lora_lexi_8b,"---
228
+ language:
229
+ - en
230
+ license: apache-2.0
231
+ tags:
232
+ - text-generation-inference
233
+ - transformers
234
+ - unsloth
235
+ - llama
236
+ - trl
237
+ base_model: Orenguteng/Llama-3-8B-Lexi-Uncensored
238
+ ---
239
+
240
+ # Uploaded model
241
+
242
+ - **Developed by:** scoliono
243
+ - **License:** apache-2.0
244
+ - **Finetuned from model :** Orenguteng/Llama-3-8B-Lexi-Uncensored
245
+
246
+ This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
247
+
248
+ [<img src=""https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png"" width=""200""/>](https://github.com/unslothai/unsloth)
249
+ ","{""id"": ""scoliono/groupchat_lora_lexi_8b"", ""author"": ""scoliono"", ""sha"": ""f133522fcee5f9264998dcba5b48443f9c50062b"", ""last_modified"": ""2024-05-28 17:14:46+00:00"", ""created_at"": ""2024-05-28 17:14:33+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""safetensors"", ""text-generation-inference"", ""unsloth"", ""llama"", ""trl"", ""en"", ""base_model:Orenguteng/Llama-3-8B-Lexi-Uncensored"", ""base_model:finetune:Orenguteng/Llama-3-8B-Lexi-Uncensored"", ""license:apache-2.0"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: Orenguteng/Llama-3-8B-Lexi-Uncensored\nlanguage:\n- en\nlicense: apache-2.0\ntags:\n- text-generation-inference\n- transformers\n- unsloth\n- llama\n- trl"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": {""auto_model"": ""AutoModel"", ""custom_class"": null, ""pipeline_tag"": null, ""processor"": null}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_model.safetensors', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-05-28 17:14:46+00:00"", ""cardData"": ""base_model: Orenguteng/Llama-3-8B-Lexi-Uncensored\nlanguage:\n- en\nlicense: apache-2.0\ntags:\n- text-generation-inference\n- transformers\n- unsloth\n- llama\n- trl"", ""transformersInfo"": {""auto_model"": ""AutoModel"", ""custom_class"": null, ""pipeline_tag"": null, ""processor"": null}, ""_id"": ""665610f9191c117e81a80314"", ""modelId"": ""scoliono/groupchat_lora_lexi_8b"", ""usedStorage"": 335604696}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=scoliono/groupchat_lora_lexi_8b&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bscoliono%2Fgroupchat_lora_lexi_8b%5D(%2Fscoliono%2Fgroupchat_lora_lexi_8b)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
250
+ canTooDdev/LlamaWalter2,"---
251
+ base_model: Orenguteng/Llama-3-8B-Lexi-Uncensored
252
+ language:
253
+ - en
254
+ license: apache-2.0
255
+ tags:
256
+ - text-generation-inference
257
+ - transformers
258
+ - unsloth
259
+ - llama
260
+ - trl
261
+ - sft
262
+ ---
263
+
264
+ # Uploaded model
265
+
266
+ - **Developed by:** canTooDdev
267
+ - **License:** apache-2.0
268
+ - **Finetuned from model :** Orenguteng/Llama-3-8B-Lexi-Uncensored
269
+
270
+ This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
271
+
272
+ [<img src=""https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png"" width=""200""/>](https://github.com/unslothai/unsloth)
273
+ ","{""id"": ""canTooDdev/LlamaWalter2"", ""author"": ""canTooDdev"", ""sha"": ""e93c01bc1fb3e1ad9e457b417c91d8edb51e82f4"", ""last_modified"": ""2024-07-18 16:26:25+00:00"", ""created_at"": ""2024-07-18 16:19:11+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 2, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""pytorch"", ""llama"", ""text-generation"", ""text-generation-inference"", ""unsloth"", ""trl"", ""sft"", ""conversational"", ""en"", ""base_model:Orenguteng/Llama-3-8B-Lexi-Uncensored"", ""base_model:finetune:Orenguteng/Llama-3-8B-Lexi-Uncensored"", ""license:apache-2.0"", ""autotrain_compatible"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: Orenguteng/Llama-3-8B-Lexi-Uncensored\nlanguage:\n- en\nlicense: apache-2.0\ntags:\n- text-generation-inference\n- transformers\n- unsloth\n- llama\n- trl\n- sft"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": null, ""config"": {""architectures"": [""LlamaForCausalLM""], ""model_type"": ""llama"", ""tokenizer_config"": {""bos_token"": ""<|begin_of_text|>"", ""chat_template"": ""{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}"", ""eos_token"": ""<|eot_id|>"", ""pad_token"": ""<|end_of_text|>""}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00001-of-00004.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00002-of-00004.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00003-of-00004.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00004-of-00004.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model.bin.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-07-18 16:26:25+00:00"", ""cardData"": ""base_model: Orenguteng/Llama-3-8B-Lexi-Uncensored\nlanguage:\n- en\nlicense: apache-2.0\ntags:\n- text-generation-inference\n- transformers\n- unsloth\n- llama\n- trl\n- sft"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""_id"": ""6699407f4bbe8ad52e7d2fef"", ""modelId"": ""canTooDdev/LlamaWalter2"", ""usedStorage"": 16060625627}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=canTooDdev/LlamaWalter2&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BcanTooDdev%2FLlamaWalter2%5D(%2FcanTooDdev%2FLlamaWalter2)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
MagicAnimate_finetunes_20250426_221535.csv_finetunes_20250426_221535.csv ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ zcxu-eric/MagicAnimate,"---
3
+ license: bsd-3-clause
4
+ ---
5
+ Download MagicAnimate checkpoints by:
6
+ ```bash
7
+ git lfs clone https://huggingface.co/zcxu-eric/MagicAnimate
8
+ ```","{""id"": ""zcxu-eric/MagicAnimate"", ""author"": ""zcxu-eric"", ""sha"": ""3d80ae8c50b289e55ee68deecc83afaab9c6a382"", ""last_modified"": ""2023-12-04 15:55:47+00:00"", ""created_at"": ""2023-12-04 15:09:36+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 186, ""library_name"": ""diffusers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""diffusers"", ""safetensors"", ""license:bsd-3-clause"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""license: bsd-3-clause"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='appearance_encoder/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='appearance_encoder/diffusion_pytorch_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='densepose_controlnet/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='densepose_controlnet/diffusion_pytorch_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='temporal_attention/temporal_attention.ckpt', size=None, blob_id=None, lfs=None)""], ""spaces"": [""zcxu-eric/magicanimate"", ""John6666/magicanimate2024"", ""cbensimon/magicanimate"", ""John6666/magicanimate"", ""Gyufyjk/magicanimate"", ""AIADC/magicanimate1"", ""walre1301/magicanimate"", ""vakilrathod67/magicanimat"", ""AIADC/magicanimate"", ""LioAlbert/magicanimate"", ""BG5/magicanimate"", ""samaleksandrov/magicanimate"", ""cgeorgia/magicanimate"", ""josh3io/magicanimate"", ""Ai3333/magicanimate"", ""AceyKubbo/magicanimate"", ""Hitendra1851/magicanimate"", ""ixlm/magicanimate"", ""neveu/magicanimate"", ""JaneMonica/magicanimate"", ""qsdreams/amimationmuse"", ""Shellbrady/magicanimate"", ""BN4350/magicanimate"", ""tomajkb/magicanimate"", ""okasf/magicanimate"", ""matejkratochvil/magicanimate"", ""jcachat/magicanimate"", ""K00B404/magicanimate""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-12-04 15:55:47+00:00"", ""cardData"": ""license: bsd-3-clause"", ""transformersInfo"": null, ""_id"": ""656debb0f7be0986b4dbb14d"", ""modelId"": ""zcxu-eric/MagicAnimate"", ""usedStorage"": 9988780119}",0,,0,,0,,0,,0,"AIADC/magicanimate, AIADC/magicanimate1, BG5/magicanimate, Gyufyjk/magicanimate, John6666/magicanimate, John6666/magicanimate2024, LioAlbert/magicanimate, cbensimon/magicanimate, huggingface/InferenceSupport/discussions/new?title=zcxu-eric/MagicAnimate&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bzcxu-eric%2FMagicAnimate%5D(%2Fzcxu-eric%2FMagicAnimate)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, samaleksandrov/magicanimate, vakilrathod67/magicanimat, walre1301/magicanimate, zcxu-eric/magicanimate",13
MeloTTS-English_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv ADDED
@@ -0,0 +1,205 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ myshell-ai/MeloTTS-English,"---
3
+ license: mit
4
+ language:
5
+ - ko
6
+ pipeline_tag: text-to-speech
7
+ ---
8
+
9
+
10
+ # MeloTTS
11
+
12
+ <a href=""https://trendshift.io/repositories/8133"" target=""_blank""><img src=""https://trendshift.io/api/badge/repositories/8133"" alt=""myshell-ai%2FMeloTTS | Trendshift"" style=""width: 250px; height: 55px;"" width=""250"" height=""55""/></a>
13
+
14
+
15
+ MeloTTS is a **high-quality multi-lingual** text-to-speech library by [MIT](https://www.mit.edu/) and [MyShell.ai](https://myshell.ai). Supported languages include:
16
+
17
+
18
+
19
+ | Model card | Example |
20
+ | --- | --- |
21
+ | [English](https://huggingface.co/myshell-ai/MeloTTS-English-v2) (American) | [Link](https://myshell-public-repo-host.s3.amazonaws.com/myshellttsbase/examples/en/EN-US/speed_1.0/sent_000.wav) |
22
+ | [English](https://huggingface.co/myshell-ai/MeloTTS-English-v2) (British) | [Link](https://myshell-public-repo-host.s3.amazonaws.com/myshellttsbase/examples/en/EN-BR/speed_1.0/sent_000.wav) |
23
+ | [English](https://huggingface.co/myshell-ai/MeloTTS-English-v2) (Indian) | [Link](https://myshell-public-repo-host.s3.amazonaws.com/myshellttsbase/examples/en/EN_INDIA/speed_1.0/sent_000.wav) |
24
+ | [English](https://huggingface.co/myshell-ai/MeloTTS-English-v2) (Australian) | [Link](https://myshell-public-repo-host.s3.amazonaws.com/myshellttsbase/examples/en/EN-AU/speed_1.0/sent_000.wav) |
25
+ | [English](https://huggingface.co/myshell-ai/MeloTTS-English-v2) (Default) | [Link](https://myshell-public-repo-host.s3.amazonaws.com/myshellttsbase/examples/en/EN-Default/speed_1.0/sent_000.wav) |
26
+ | [Spanish](https://huggingface.co/myshell-ai/MeloTTS-Spanish) | [Link](https://myshell-public-repo-host.s3.amazonaws.com/myshellttsbase/examples/es/ES/speed_1.0/sent_000.wav) |
27
+ | [French](https://huggingface.co/myshell-ai/MeloTTS-French) | [Link](https://myshell-public-repo-host.s3.amazonaws.com/myshellttsbase/examples/fr/FR/speed_1.0/sent_000.wav) |
28
+ | [Chinese](https://huggingface.co/myshell-ai/MeloTTS-Chinese) (mix EN) | [Link](https://myshell-public-repo-host.s3.amazonaws.com/myshellttsbase/examples/zh/ZH/speed_1.0/sent_008.wav) |
29
+ | [Japanese](https://huggingface.co/myshell-ai/MeloTTS-Japanese) | [Link](https://myshell-public-repo-host.s3.amazonaws.com/myshellttsbase/examples/jp/JP/speed_1.0/sent_000.wav) |
30
+ | [Korean](https://huggingface.co/myshell-ai/MeloTTS-Korean/) | [Link](https://myshell-public-repo-host.s3.amazonaws.com/myshellttsbase/examples/kr/KR/speed_1.0/sent_000.wav) |
31
+
32
+ Some other features include:
33
+ - The Chinese speaker supports `mixed Chinese and English`.
34
+ - Fast enough for `CPU real-time inference`.
35
+
36
+ ## Authors
37
+
38
+ - [Wenliang Zhao](https://wl-zhao.github.io) at Tsinghua University
39
+ - [Xumin Yu](https://yuxumin.github.io) at Tsinghua University
40
+ - [Zengyi Qin](https://www.qinzy.tech) (project lead) at MIT and MyShell
41
+
42
+ **Citation**
43
+ ```
44
+ @software{zhao2024melo,
45
+ author={Zhao, Wenliang and Yu, Xumin and Qin, Zengyi},
46
+ title = {MeloTTS: High-quality Multi-lingual Multi-accent Text-to-Speech},
47
+ url = {https://github.com/myshell-ai/MeloTTS},
48
+ year = {2023}
49
+ }
50
+ ```
51
+
52
+
53
+ ## Usage
54
+
55
+ ### Without Installation
56
+
57
+ An unofficial [live demo](https://huggingface.co/spaces/mrfakename/MeloTTS) is hosted on Hugging Face Spaces.
58
+
59
+ #### Use it on MyShell
60
+
61
+ There are hundreds of TTS models on MyShell, much more than MeloTTS. See examples [here](https://github.com/myshell-ai/MeloTTS/blob/main/docs/quick_use.md#use-melotts-without-installation).
62
+ More can be found at the widget center of [MyShell.ai](https://app.myshell.ai/robot-workshop).
63
+
64
+ ### Install and Use Locally
65
+
66
+ Follow the installation steps [here](https://github.com/myshell-ai/MeloTTS/blob/main/docs/install.md#linux-and-macos-install) before using the following snippet:
67
+
68
+ ```python
69
+ from melo.api import TTS
70
+
71
+ # Speed is adjustable
72
+ speed = 1.0
73
+
74
+ # CPU is sufficient for real-time inference.
75
+ # You can set it manually to 'cpu' or 'cuda' or 'cuda:0' or 'mps'
76
+ device = 'auto' # Will automatically use GPU if available
77
+
78
+ # English
79
+ text = ""Did you ever hear a folk tale about a giant turtle?""
80
+ model = TTS(language='EN', device=device)
81
+ speaker_ids = model.hps.data.spk2id
82
+
83
+ # American accent
84
+ output_path = 'en-us.wav'
85
+ model.tts_to_file(text, speaker_ids['EN-US'], output_path, speed=speed)
86
+
87
+ # British accent
88
+ output_path = 'en-br.wav'
89
+ model.tts_to_file(text, speaker_ids['EN-BR'], output_path, speed=speed)
90
+
91
+ # Indian accent
92
+ output_path = 'en-india.wav'
93
+ model.tts_to_file(text, speaker_ids['EN_INDIA'], output_path, speed=speed)
94
+
95
+ # Australian accent
96
+ output_path = 'en-au.wav'
97
+ model.tts_to_file(text, speaker_ids['EN-AU'], output_path, speed=speed)
98
+
99
+ # Default accent
100
+ output_path = 'en-default.wav'
101
+ model.tts_to_file(text, speaker_ids['EN-Default'], output_path, speed=speed)
102
+
103
+ ```
104
+
105
+
106
+ ## Join the Community
107
+
108
+ **Open Source AI Grant**
109
+
110
+ We are actively sponsoring open-source AI projects. The sponsorship includes GPU resources, fundings and intellectual support (collaboration with top research labs). We welcome both reseach and engineering projects, as long as the open-source community needs them. Please contact [Zengyi Qin](https://www.qinzy.tech/) if you are interested.
111
+
112
+ **Contributing**
113
+
114
+ If you find this work useful, please consider contributing to the GitHub [repo](https://github.com/myshell-ai/MeloTTS).
115
+
116
+ - Many thanks to [@fakerybakery](https://github.com/fakerybakery) for adding the Web UI and CLI part.
117
+
118
+ ## License
119
+
120
+ This library is under MIT License, which means it is free for both commercial and non-commercial use.
121
+
122
+ ## Acknowledgements
123
+
124
+ This implementation is based on [TTS](https://github.com/coqui-ai/TTS), [VITS](https://github.com/jaywalnut310/vits), [VITS2](https://github.com/daniilrobnikov/vits2) and [Bert-VITS2](https://github.com/fishaudio/Bert-VITS2). We appreciate their awesome work.
125
+
126
+ ","{""id"": ""myshell-ai/MeloTTS-English"", ""author"": ""myshell-ai"", ""sha"": ""bb4fb7346d566d277ba8c8c7dbfdf6786139b8ef"", ""last_modified"": ""2024-12-24 19:18:31+00:00"", ""created_at"": ""2024-02-29 14:52:43+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 212802, ""downloads_all_time"": null, ""likes"": 265, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""text-to-speech"", ""ko"", ""license:mit"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""text-to-speech"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""language:\n- ko\nlicense: mit\npipeline_tag: text-to-speech"", ""widget_data"": null, ""model_index"": null, ""config"": {}, ""transformers_info"": {""auto_model"": ""AutoModel"", ""custom_class"": null, ""pipeline_tag"": null, ""processor"": null}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='checkpoint.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [""gabrielchua/open-notebooklm"", ""mrfakename/MeloTTS"", ""neuromod0/MeloTTS-English-v3"", ""reflex-ai/MeloTTS-English-v3"", ""subhasmita/myshell-ai-MeloTTS-English"", ""ishworrsubedii/MeloTTS"", ""derivativegenius/myshell-ai-MeloTTS-English"", ""vuxuanhoan/MeloTTS"", ""Noe831/tts"", ""SupariKoli/myshell-ai-MeloTTS-English"", ""Ahmed0011/myshell-ai-MeloTTS-English"", ""Memoroeisdead/myshell-ai-MeloTTS-English"", ""Mazen001/myshell-ai-MeloTTS-English"", ""Jacobsirdan/myshell-ai-MeloTTS-English"", ""atlasias/myshell-ai-MeloTTS-English"", ""Pavan178/myshell-ai-MeloTTS-English"", ""zetabyte/myshell-ai-MeloTTS-English"", ""Dima123e/myshell-ai-MeloTTS-English"", ""ReySajju742/myshell-ai-MeloTTS-English"", ""SerSleepy/myshell-ai-MeloTTS-English"", ""dijj/myshell-ai-MeloTTS-English"", ""creospin/myshell-ai-MeloTTS-English"", ""ssenwshj/myshell-ai-MeloTTS-English"", ""CDOM201/domtts"", ""ishaank123/myshell-ai-MeloTTS-English"", ""Emienent/myshell-ai-MeloTTS-English"", ""LongTran1996/myshell-ai-MeloTTS-English"", ""SalmanAhmad-24/myshell-ai-MeloTTS-English"", ""jiuzhou223/myshell-ai-MeloTTS-English"", ""doctumdoces/myshell-ai-MeloTTS-English"", ""sumittechmero/MeloTTS"", ""Shangkhonil/Image_TO_Speech"", ""abis90/myshell-ai-MeloTTS-English"", ""abis90/myshell-ai-MeloTTS-Englishs"", ""nirajandhakal/MeloTTS"", ""dbarks/open-notebooklm"", ""cybercody/open-notebooklm"", ""TerryZazu/myshell-ai-MeloTTS-English"", ""cls7908/open-notebooklm"", ""iukea/open-notebooklm"", ""zohairy/open-notebooklm"", ""cagiraudo56789/open-notebooklm"", ""neuromod0/open-notebooklm"", ""AI-Platform/open-notebooklm"", ""WodeDadao/open-notebooklm"", ""addyosmani/open-notebooklm"", ""Pawitt/myshell"", ""slkreddy/open-notebooklm"", ""wagnergod/open-notebooklm"", ""LeonEr/myshell-ai-MeloTTS-English"", ""mahunyu66/myshell-ai-MeloTTS-English"", ""vismaya2939/textTOspeech"", ""xrainxshadowx/myshell-ai-MeloTTS-English"", ""Telistra/myshell-ai-MeloTTS-English"", ""keshav6936/myshell-ai-MeloTTS-English"", ""Rajsinghfanfjg/myshell-ai-MeloTTS-English"", ""shrolr/myshell-ai-MeloTTS-English"", ""soiz/myshell-ai-MeloTTS-English"", ""Neear1337/myshell-ai-MeloTTS-English"", ""Neear1337/MeloTTS"", ""KalaiyarasanJacob/myshell-ai-MeloTTS-English"", ""zlillymp/myshell-ai-MeloTTS-English"", ""mukeshkr5/myshell-ai-MeloTTS-English"", ""mxyzplk/myshell-ai-MeloTTS-English"", ""fullstuckdev/myshell-ai-MeloTTS-English"", ""blind1234/myshell-ai-MeloTTS-English"", ""mnisham/myshell-ai-MeloTTS-English"", ""Shamlan321/myshell-ai-MeloTTS-English"", ""hrsprojects/myshell-ai-MeloTTS-English"", ""utkubulkan/MeloTTS"", ""RORONovaLuffy/myshell-ai-MeloTTS-English"", ""huysynf/myshell-ai-MeloTTS-English"", ""rockerritesh/myshell-ai-MeloTTS-English""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-12-24 19:18:31+00:00"", ""cardData"": ""language:\n- ko\nlicense: mit\npipeline_tag: text-to-speech"", ""transformersInfo"": {""auto_model"": ""AutoModel"", ""custom_class"": null, ""pipeline_tag"": null, ""processor"": null}, ""_id"": ""65e09a3b51f0487385110f72"", ""modelId"": ""myshell-ai/MeloTTS-English"", ""usedStorage"": 207860748}",0,https://huggingface.co/kadirnar/melotts-jenny,1,,0,,0,,0,"Noe831/tts, derivativegenius/myshell-ai-MeloTTS-English, gabrielchua/open-notebooklm, hrsprojects/myshell-ai-MeloTTS-English, huggingface/InferenceSupport/discussions/new?title=myshell-ai/MeloTTS-English&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bmyshell-ai%2FMeloTTS-English%5D(%2Fmyshell-ai%2FMeloTTS-English)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, ishworrsubedii/MeloTTS, mrfakename/MeloTTS, neuromod0/MeloTTS-English-v3, reflex-ai/MeloTTS-English-v3, soiz/myshell-ai-MeloTTS-English, subhasmita/myshell-ai-MeloTTS-English, sumittechmero/MeloTTS, vuxuanhoan/MeloTTS",13
127
+ kadirnar/melotts-jenny,"---
128
+ license: mit
129
+ datasets:
130
+ - reach-vb/jenny_tts_dataset
131
+ language:
132
+ - en
133
+ base_model:
134
+ - myshell-ai/MeloTTS-English
135
+ tags:
136
+ - audio
137
+ - melotts
138
+ ---
139
+ # MeloTTS Model Checkpoint
140
+
141
+ This repository contains trained model checkpoints for MeloTTS, a high-quality multi-lingual text-to-speech system. These checkpoints are part of a trained model that can be used for text-to-speech synthesis.
142
+
143
+ ## Model Details
144
+
145
+ - **Model Type**: MeloTTS
146
+ - **Language Support**: English (Default)
147
+ - **Sampling Rate**: 44.1kHz
148
+ - **Mel Channels**: 128
149
+ - **Hidden Channels**: 192
150
+ - **Filter Channels**: 768
151
+
152
+ ### Architecture Details
153
+ - Inter channels: 192
154
+ - Number of heads: 2
155
+ - Number of layers: 6
156
+ - Flow layers: 3
157
+ - Kernel size: 3
158
+ - Dropout rate: 0.1
159
+
160
+ ## Training Dataset
161
+
162
+ This model was trained on the [Jenny TTS Dataset](https://huggingface.co/datasets/reach-vb/jenny_tts_dataset), which is available on Hugging Face. The dataset consists of high-quality English speech recordings suitable for text-to-speech training.
163
+
164
+ ## Model Files
165
+
166
+ The repository contains several checkpoint files:
167
+ - `DUR_*.pth`: Duration predictor checkpoints
168
+ - `G_*.pth`: Generator model checkpoints
169
+ - `D_*.pth`: Discriminator model checkpoints
170
+ - `config.json`: Model configuration file
171
+
172
+ ## Usage
173
+
174
+ To use this model with MeloTTS:
175
+
176
+ ```python
177
+ from melo.api import TTS
178
+
179
+ # Initialize TTS with the model path
180
+ tts = TTS(model_path=""kadirnar/melotts-model"")
181
+
182
+ # Generate speech
183
+ tts.tts_to_file(
184
+ text=""Your text here"",
185
+ speaker=""EN-default"",
186
+ language=""EN"",
187
+ output_path=""output.wav""
188
+ )
189
+ ```
190
+
191
+ ## Training Details
192
+
193
+ The model was trained with the following specifications:
194
+ - Batch size: 6
195
+ - Learning rate: 0.0003
196
+ - Beta values: [0.8, 0.99]
197
+ - Segment size: 16384
198
+
199
+ ## Original Repository
200
+
201
+ This model is based on [MeloTTS](https://github.com/myshell-ai/MeloTTS) by MyShell.ai. Visit the original repository for more details about the architecture and implementation.
202
+
203
+ ## License
204
+
205
+ This model follows the same licensing as the original MeloTTS repository (MIT License).","{""id"": ""kadirnar/melotts-jenny"", ""author"": ""kadirnar"", ""sha"": ""88630bd418b92abb54c8064e2fb26202020b6728"", ""last_modified"": ""2024-11-28 07:16:04+00:00"", ""created_at"": ""2024-11-28 07:05:03+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 2, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""tensorboard"", ""audio"", ""melotts"", ""en"", ""dataset:reach-vb/jenny_tts_dataset"", ""base_model:myshell-ai/MeloTTS-English"", ""base_model:finetune:myshell-ai/MeloTTS-English"", ""license:mit"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- myshell-ai/MeloTTS-English\ndatasets:\n- reach-vb/jenny_tts_dataset\nlanguage:\n- en\nlicense: mit\ntags:\n- audio\n- melotts"", ""widget_data"": null, ""model_index"": null, ""config"": {}, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DUR_0.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DUR_77000.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DUR_78000.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='DUR_79000.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='D_0.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='D_78000.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='D_79000.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='G_0.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='G_75000.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='G_76000.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='G_77000.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='G_78000.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='G_79000.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='eval/events.out.tfevents.1732646944.optimistic-hubble.1174083.1', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='eval/events.out.tfevents.1732656741.optimistic-hubble.1237661.1', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='eval/events.out.tfevents.1732692904.optimistic-hubble.1294737.1', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='eval/events.out.tfevents.1732715645.optimistic-hubble.1345340.1', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='eval/events.out.tfevents.1732720979.optimistic-hubble.1367391.1', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='eval/events.out.tfevents.1732733258.optimistic-hubble.1406475.1', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='eval/events.out.tfevents.1732733614.optimistic-hubble.1417031.1', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='eval/events.out.tfevents.1732739659.optimistic-hubble.1440705.1', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='eval/events.out.tfevents.1732743102.optimistic-hubble.1460255.1', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='eval/events.out.tfevents.1732747168.optimistic-hubble.1478394.1', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='eval/events.out.tfevents.1732776627.optimistic-hubble.1518651.1', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='events.out.tfevents.1732646944.optimistic-hubble.1174083.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='events.out.tfevents.1732656741.optimistic-hubble.1237661.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='events.out.tfevents.1732692904.optimistic-hubble.1294737.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='events.out.tfevents.1732715645.optimistic-hubble.1345340.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='events.out.tfevents.1732720979.optimistic-hubble.1367391.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='events.out.tfevents.1732733258.optimistic-hubble.1406475.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='events.out.tfevents.1732733614.optimistic-hubble.1417031.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='events.out.tfevents.1732739659.optimistic-hubble.1440705.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='events.out.tfevents.1732743102.optimistic-hubble.1460255.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='events.out.tfevents.1732747168.optimistic-hubble.1478394.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='events.out.tfevents.1732776627.optimistic-hubble.1518651.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='train.log', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-11-28 07:16:04+00:00"", ""cardData"": ""base_model:\n- myshell-ai/MeloTTS-English\ndatasets:\n- reach-vb/jenny_tts_dataset\nlanguage:\n- en\nlicense: mit\ntags:\n- audio\n- melotts"", ""transformersInfo"": null, ""_id"": ""6748161f102b62dffc6bb9ea"", ""modelId"": ""kadirnar/melotts-jenny"", ""usedStorage"": 5871054267}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=kadirnar/melotts-jenny&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bkadirnar%2Fmelotts-jenny%5D(%2Fkadirnar%2Fmelotts-jenny)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
MiniCPM-Llama3-V-2_5-gguf_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ openbmb/MiniCPM-Llama3-V-2_5-gguf,"---
3
+ tags:
4
+ - llama.cpp
5
+ base_model:
6
+ - openbmb/MiniCPM-Llama3-V-2_5
7
+ ---
8
+
9
+ # MiniCPM-Llama3-V 2.5 gguf files for llama.cpp
10
+
11
+ ## Usage
12
+ Please see our fork of [llama.cpp](https://github.com/OpenBMB/llama.cpp/tree/minicpm-v2.5/examples/minicpmv) for more detail to run MiniCPM-Llama3-V 2.5 with llama.cpp
13
+
14
+ ## ollama
15
+ [ollama](https://github.com/OpenBMB/ollama/tree/minicpm-v2.5/examples/minicpm-v2.5)","{""id"": ""openbmb/MiniCPM-Llama3-V-2_5-gguf"", ""author"": ""openbmb"", ""sha"": ""d090fbfffeef9f01dfb282b3cf2b1e3360dfd29a"", ""last_modified"": ""2025-02-27 08:22:43+00:00"", ""created_at"": ""2024-05-19 17:35:26+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 8138, ""downloads_all_time"": null, ""likes"": 213, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""gguf"", ""llama.cpp"", ""base_model:openbmb/MiniCPM-Llama3-V-2_5"", ""base_model:quantized:openbmb/MiniCPM-Llama3-V-2_5"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- openbmb/MiniCPM-Llama3-V-2_5\ntags:\n- llama.cpp"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-BF16.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-F16.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-IQ3_M.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-IQ3_S.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-IQ3_XS.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-IQ4_NL.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-IQ4_XS.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q2_K.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q3_K.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q3_K_L.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q3_K_M.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q3_K_S.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q4_0.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q4_1.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q4_K.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q4_K_M.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q4_K_M_for_pr.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q4_K_S.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q5_0.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q5_1.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q5_K.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q5_K_M.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q5_K_S.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q6_K.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q8_0.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='mmproj-model-f16.gguf', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-27 08:22:43+00:00"", ""cardData"": ""base_model:\n- openbmb/MiniCPM-Llama3-V-2_5\ntags:\n- llama.cpp"", ""transformersInfo"": null, ""_id"": ""664a385e7a1ed3e001f3d8f7"", ""modelId"": ""openbmb/MiniCPM-Llama3-V-2_5-gguf"", ""usedStorage"": 139601373248}",0,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=openbmb/MiniCPM-Llama3-V-2_5-gguf&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bopenbmb%2FMiniCPM-Llama3-V-2_5-gguf%5D(%2Fopenbmb%2FMiniCPM-Llama3-V-2_5-gguf)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
MiniCPM-V-2_6-gguf_finetunes_20250427_003734.csv_finetunes_20250427_003734.csv ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ openbmb/MiniCPM-V-2_6-gguf,"---
3
+ base_model:
4
+ - openbmb/MiniCPM-V-2_6
5
+ ---
6
+ ## MiniCPM-V 2.6
7
+
8
+ ### Prepare models and code
9
+
10
+ Download [MiniCPM-V-2_6](https://huggingface.co/openbmb/MiniCPM-V-2_6) PyTorch model from huggingface to ""MiniCPM-V-2_6"" folder.
11
+
12
+ Clone llama.cpp:
13
+ ```bash
14
+ git clone git@github.com:OpenBMB/llama.cpp.git
15
+ cd llama.cpp
16
+ git checkout minicpmv-main
17
+ ```
18
+
19
+ ### Usage of MiniCPM-V 2.6
20
+
21
+ Convert PyTorch model to gguf files (You can also download the converted [gguf](https://huggingface.co/openbmb/MiniCPM-V-2_6-gguf) by us)
22
+
23
+ ```bash
24
+ python ./examples/llava/minicpmv-surgery.py -m ../MiniCPM-V-2_6
25
+ python ./examples/llava/minicpmv-convert-image-encoder-to-gguf.py -m ../MiniCPM-V-2_6 --minicpmv-projector ../MiniCPM-V-2_6/minicpmv.projector --output-dir ../MiniCPM-V-2_6/ --image-mean 0.5 0.5 0.5 --image-std 0.5 0.5 0.5 --minicpmv_version 3
26
+ python ./convert_hf_to_gguf.py ../MiniCPM-V-2_6/model
27
+
28
+ # quantize int4 version
29
+ ./llama-quantize ../MiniCPM-V-2_6/model/ggml-model-f16.gguf ../MiniCPM-V-2_6/model/ggml-model-Q4_K_M.gguf Q4_K_M
30
+ ```
31
+
32
+ Build for Linux or Mac
33
+
34
+ ```bash
35
+ make
36
+ make llama-minicpmv-cli
37
+ ```
38
+
39
+ Inference on Linux or Mac
40
+ ```
41
+ # run f16 version
42
+ ./llama-minicpmv-cli -m ../MiniCPM-V-2_6/model/ggml-model-f16.gguf --mmproj ../MiniCPM-V-2_6/mmproj-model-f16.gguf -c 4096 --temp 0.7 --top-p 0.8 --top-k 100 --repeat-penalty 1.05 --image xx.jpg -p ""What is in the image?""
43
+
44
+ # run quantized int4 version
45
+ ./llama-minicpmv-cli -m ../MiniCPM-V-2_6/model/ggml-model-Q4_K_M.gguf --mmproj ../MiniCPM-V-2_6/mmproj-model-f16.gguf -c 4096 --temp 0.7 --top-p 0.8 --top-k 100 --repeat-penalty 1.05 --image xx.jpg -p ""What is in the image?""
46
+
47
+ # or run in interactive mode
48
+ ./llama-minicpmv-cli -m ../MiniCPM-V-2_6/model/ggml-model-Q4_K_M.gguf --mmproj ../MiniCPM-V-2_6/mmproj-model-f16.gguf -c 4096 --temp 0.7 --top-p 0.8 --top-k 100 --repeat-penalty 1.05 --image xx.jpg -i
49
+ ```","{""id"": ""openbmb/MiniCPM-V-2_6-gguf"", ""author"": ""openbmb"", ""sha"": ""48fe6436abf57b3df6ec34f73cdc1fb4b740acb0"", ""last_modified"": ""2025-02-27 08:22:16+00:00"", ""created_at"": ""2024-08-04 05:55:37+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 19692, ""downloads_all_time"": null, ""likes"": 159, ""library_name"": null, ""gguf"": {""total"": 7612763648, ""architecture"": ""qwen2"", ""context_length"": 32768, ""chat_template"": ""{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}"", ""bos_token"": ""<|im_start|>"", ""eos_token"": ""<|im_end|>""}, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""gguf"", ""base_model:openbmb/MiniCPM-V-2_6"", ""base_model:quantized:openbmb/MiniCPM-V-2_6"", ""endpoints_compatible"", ""region:us"", ""conversational""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- openbmb/MiniCPM-V-2_6"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-IQ3_M.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-IQ3_S.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-IQ3_XS.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-IQ4_NL.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-IQ4_XS.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q2_K.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q3_K.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q3_K_L.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q3_K_M.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q3_K_S.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q4_0.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q4_1.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q4_K.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q4_K_M.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q4_K_S.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q5_0.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q5_1.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q5_K.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q5_K_M.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q5_K_S.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q6_K.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-Q8_0.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ggml-model-f16.gguf', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='mmproj-model-f16.gguf', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-27 08:22:16+00:00"", ""cardData"": ""base_model:\n- openbmb/MiniCPM-V-2_6"", ""transformersInfo"": null, ""_id"": ""66af17d93c257dab06a05104"", ""modelId"": ""openbmb/MiniCPM-V-2_6-gguf"", ""usedStorage"": 105451169760}",0,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=openbmb/MiniCPM-V-2_6-gguf&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bopenbmb%2FMiniCPM-V-2_6-gguf%5D(%2Fopenbmb%2FMiniCPM-V-2_6-gguf)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
Molmo-7B-D-0924_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv ADDED
@@ -0,0 +1,211 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ allenai/Molmo-7B-D-0924,"---
3
+ license: apache-2.0
4
+ language:
5
+ - en
6
+ base_model:
7
+ - openai/clip-vit-large-patch14-336
8
+ - Qwen/Qwen2-7B
9
+ pipeline_tag: image-text-to-text
10
+ tags:
11
+ - multimodal
12
+ - olmo
13
+ - molmo
14
+ - pixmo
15
+ library_name: transformers
16
+ ---
17
+
18
+ <img src=""molmo_logo.png"" alt=""Logo for the Molmo Project"" style=""width: auto; height: 50px;"">
19
+
20
+ # Molmo 7B-D
21
+
22
+ Molmo is a family of open vision-language models developed by the Allen Institute for AI. Molmo models are trained on PixMo, a dataset of 1 million, highly-curated image-text pairs. It has state-of-the-art performance among multimodal models with a similar size while being fully open-source. You can find all models in the Molmo family [here](https://huggingface.co/collections/allenai/molmo-66f379e6fe3b8ef090a8ca19).
23
+ **Learn more** about the Molmo family [in our announcement blog post](https://molmo.allenai.org/blog) or the [paper](https://huggingface.co/papers/2409.17146).
24
+
25
+ Molmo 7B-D is based on [Qwen2-7B](https://huggingface.co/Qwen/Qwen2-7B) and uses [OpenAI CLIP](https://huggingface.co/openai/clip-vit-large-patch14-336) as vision backbone.
26
+ It performs comfortably between GPT-4V and GPT-4o on both academic benchmarks and human evaluation.
27
+ It powers the **Molmo demo at** [**molmo.allenai.org**](https://molmo.allenai.org).
28
+
29
+ This checkpoint is a **preview** of the Molmo release. All artifacts used in creating Molmo (PixMo dataset, training code, evaluations, intermediate checkpoints) will be made available at a later date, furthering our commitment to open-source AI development and reproducibility.
30
+
31
+ [**Sign up here**](https://docs.google.com/forms/d/e/1FAIpQLSdML1MhNNBDsCHpgWG65Oydg2SjZzVasyqlP08nBrWjZp_c7A/viewform) to be the first to know when artifacts are released.
32
+
33
+ Quick links:
34
+ - 💬 [Demo](https://molmo.allenai.org/)
35
+ - 📂 [All Models](https://huggingface.co/collections/allenai/molmo-66f379e6fe3b8ef090a8ca19)
36
+ - 📃 [Paper](https://molmo.allenai.org/paper.pdf)
37
+ - 🎥 [Blog with Videos](https://molmo.allenai.org/blog)
38
+
39
+
40
+ ## Quick Start
41
+
42
+ To run Molmo, first install dependencies:
43
+
44
+ ```bash
45
+ pip install einops torchvision
46
+ ```
47
+
48
+ Then, follow these steps:
49
+
50
+ ```python
51
+ from transformers import AutoModelForCausalLM, AutoProcessor, GenerationConfig
52
+ from PIL import Image
53
+ import requests
54
+
55
+ # load the processor
56
+ processor = AutoProcessor.from_pretrained(
57
+ 'allenai/Molmo-7B-D-0924',
58
+ trust_remote_code=True,
59
+ torch_dtype='auto',
60
+ device_map='auto'
61
+ )
62
+
63
+ # load the model
64
+ model = AutoModelForCausalLM.from_pretrained(
65
+ 'allenai/Molmo-7B-D-0924',
66
+ trust_remote_code=True,
67
+ torch_dtype='auto',
68
+ device_map='auto'
69
+ )
70
+
71
+ # process the image and text
72
+ inputs = processor.process(
73
+ images=[Image.open(requests.get(""https://picsum.photos/id/237/536/354"", stream=True).raw)],
74
+ text=""Describe this image.""
75
+ )
76
+
77
+ # move inputs to the correct device and make a batch of size 1
78
+ inputs = {k: v.to(model.device).unsqueeze(0) for k, v in inputs.items()}
79
+
80
+ # generate output; maximum 200 new tokens; stop generation when <|endoftext|> is generated
81
+ output = model.generate_from_batch(
82
+ inputs,
83
+ GenerationConfig(max_new_tokens=200, stop_strings=""<|endoftext|>""),
84
+ tokenizer=processor.tokenizer
85
+ )
86
+
87
+ # only get generated tokens; decode them to text
88
+ generated_tokens = output[0,inputs['input_ids'].size(1):]
89
+ generated_text = processor.tokenizer.decode(generated_tokens, skip_special_tokens=True)
90
+
91
+ # print the generated text
92
+ print(generated_text)
93
+
94
+ # >>> This image features an adorable black Labrador puppy, captured from a top-down
95
+ # perspective. The puppy is sitting on a wooden deck, which is composed ...
96
+ ```
97
+
98
+ To make inference more efficient, run with autocast:
99
+
100
+ ```python
101
+ with torch.autocast(device_type=""cuda"", enabled=True, dtype=torch.bfloat16):
102
+ output = model.generate_from_batch(
103
+ inputs,
104
+ GenerationConfig(max_new_tokens=200, stop_strings=""<|endoftext|>""),
105
+ tokenizer=processor.tokenizer
106
+ )
107
+ ```
108
+
109
+ We did most of our evaluation in this setting (autocast on, but float32 weights)
110
+
111
+ To even further reduce the memory requirements, the model can be run with bfloat16 weights:
112
+
113
+ ```python
114
+ model.to(dtype=torch.bfloat16)
115
+ inputs[""images""] = inputs[""images""].to(torch.bfloat16)
116
+ output = model.generate_from_batch(
117
+ inputs,
118
+ GenerationConfig(max_new_tokens=200, stop_strings=""<|endoftext|>""),
119
+ tokenizer=processor.tokenizer
120
+ )
121
+ ```
122
+
123
+ Note that we have observed that this can change the output of the model compared to running with float32 weights.
124
+
125
+ ## Evaluations
126
+
127
+ | Model | Average Score on 11 Academic Benchmarks | Human Preference Elo Rating |
128
+ |-----------------------------|-----------------------------------------|-----------------------------|
129
+ | Molmo 72B | 81.2 | 1077 |
130
+ | **Molmo 7B-D (this model)** | **77.3** | **1056** |
131
+ | Molmo 7B-O | 74.6 | 1051 |
132
+ | MolmoE 1B | 68.6 | 1032 |
133
+ | GPT-4o | 78.5 | 1079 |
134
+ | GPT-4V | 71.1 | 1041 |
135
+ | Gemini 1.5 Pro | 78.3 | 1074 |
136
+ | Gemini 1.5 Flash | 75.1 | 1054 |
137
+ | Claude 3.5 Sonnet | 76.7 | 1069 |
138
+ | Claude 3 Opus | 66.4 | 971 |
139
+ | Claude 3 Haiku | 65.3 | 999 |
140
+ | Qwen VL2 72B | 79.4 | 1037 |
141
+ | Qwen VL2 7B | 73.7 | 1025 |
142
+ | Intern VL2 LLAMA 76B | 77.1 | 1018 |
143
+ | Intern VL2 8B | 69.4 | 953 |
144
+ | Pixtral 12B | 69.5 | 1016 |
145
+ | Phi3.5-Vision 4B | 59.7 | 982 |
146
+ | PaliGemma 3B | 50.0 | 937 |
147
+ | LLAVA OneVision 72B | 76.6 | 1051 |
148
+ | LLAVA OneVision 7B | 72.0 | 1024 |
149
+ | Cambrian-1 34B | 66.8 | 953 |
150
+ | Cambrian-1 8B | 63.4 | 952 |
151
+ | xGen - MM - Interleave 4B | 59.5 | 979 |
152
+ | LLAVA-1.5 13B | 43.9 | 960 |
153
+ | LLAVA-1.5 7B | 40.7 | 951 |
154
+
155
+ *Benchmarks: AI2D test, ChartQA test, VQA v2.0 test, DocQA test, InfographicVQA test, TextVQA val, RealWorldQA, MMMU val, MathVista testmini, CountBenchQA, Flickr Count (we collected this new dataset that is significantly harder than CountBenchQA).*
156
+
157
+ ## FAQs
158
+
159
+ ### I'm getting an error a broadcast error when processing images!
160
+
161
+ Your image might not be in RGB format. You can convert it using the following code snippet:
162
+
163
+ ```python
164
+ from PIL import Image
165
+
166
+ image = Image.open(...)
167
+
168
+ if image.mode != ""RGB"":
169
+ image = image.convert(""RGB"")
170
+ ```
171
+
172
+ ### Molmo doesn't work great with transparent images!
173
+
174
+ We received reports that Molmo models might struggle with transparent images.
175
+ For the time being, we recommend adding a white or dark background to your images before passing them to the model. The code snippet below shows how to do this using the Python Imaging Library (PIL):
176
+
177
+ ```python
178
+
179
+ # Load the image
180
+ url = ""...""
181
+ image = Image.open(requests.get(url, stream=True).raw)
182
+
183
+ # Convert the image to grayscale to calculate brightness
184
+ gray_image = image.convert('L') # Convert to grayscale
185
+
186
+ # Calculate the average brightness
187
+ stat = ImageStat.Stat(gray_image)
188
+ average_brightness = stat.mean[0] # Get the average value
189
+
190
+ # Define background color based on brightness (threshold can be adjusted)
191
+ bg_color = (0, 0, 0) if average_brightness > 127 else (255, 255, 255)
192
+
193
+ # Create a new image with the same size as the original, filled with the background color
194
+ new_image = Image.new('RGB', image.size, bg_color)
195
+
196
+ # Paste the original image on top of the background (use image as a mask if needed)
197
+ new_image.paste(image, (0, 0), image if image.mode == 'RGBA' else None)
198
+
199
+ # Now you can pass the new_image to Molmo
200
+ processor = AutoProcessor.from_pretrained(
201
+ 'allenai/Molmo-7B-D-0924',
202
+ trust_remote_code=True,
203
+ torch_dtype='auto',
204
+ device_map='auto'
205
+ )
206
+ ```
207
+
208
+ ## License and Use
209
+
210
+ This model is licensed under Apache 2.0. It is intended for research and educational use.
211
+ For more information, please see our [Responsible Use Guidelines](https://allenai.org/responsible-use).","{""id"": ""allenai/Molmo-7B-D-0924"", ""author"": ""allenai"", ""sha"": ""ac032b93b84a7f10c9578ec59f9f20ee9a8990a2"", ""last_modified"": ""2025-04-04 20:01:44+00:00"", ""created_at"": ""2024-09-25 01:48:22+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 569560, ""downloads_all_time"": null, ""likes"": 524, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""safetensors"", ""molmo"", ""text-generation"", ""multimodal"", ""olmo"", ""pixmo"", ""image-text-to-text"", ""conversational"", ""custom_code"", ""en"", ""arxiv:2409.17146"", ""base_model:Qwen/Qwen2-7B"", ""base_model:finetune:Qwen/Qwen2-7B"", ""license:apache-2.0"", ""autotrain_compatible"", ""region:us""], ""pipeline_tag"": ""image-text-to-text"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- openai/clip-vit-large-patch14-336\n- Qwen/Qwen2-7B\nlanguage:\n- en\nlibrary_name: transformers\nlicense: apache-2.0\npipeline_tag: image-text-to-text\ntags:\n- multimodal\n- olmo\n- molmo\n- pixmo"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": null, ""config"": {""architectures"": [""MolmoForCausalLM""], ""auto_map"": {""AutoConfig"": ""config_molmo.MolmoConfig"", ""AutoModelForCausalLM"": ""modeling_molmo.MolmoForCausalLM""}, ""model_type"": ""molmo"", ""tokenizer_config"": {""bos_token"": null, ""chat_template"": ""{% for message in messages -%}\n {%- if (loop.index % 2 == 1 and message['role'] != 'user') or \n (loop.index % 2 == 0 and message['role'].lower() != 'assistant') -%}\n {{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}\n {%- endif -%}\n {{ message['role'].capitalize() + ': ' + message['content'] }}\n {%- if not loop.last -%}\n {{ ' ' }}\n {%- endif %}\n {%- endfor -%}\n {%- if add_generation_prompt -%}\n {{ ' Assistant:' }}\n {%- endif %}"", ""eos_token"": ""<|endoftext|>"", ""pad_token"": ""<|endoftext|>"", ""unk_token"": null}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": ""modeling_molmo.MolmoForCausalLM"", ""pipeline_tag"": ""text-generation"", ""processor"": null}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='added_tokens.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config_molmo.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image_preprocessing_molmo.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='merges.txt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00001-of-00007.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00002-of-00007.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00003-of-00007.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00004-of-00007.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00005-of-00007.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00006-of-00007.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00007-of-00007.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='modeling_molmo.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='molmo_logo.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='preprocessing_molmo.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='processor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='vocab.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [""akhaliq/Molmo-7B-D-0924"", ""TIGER-Lab/MEGA-Bench"", ""KBaba7/Quant"", ""bhaskartripathi/LLM_Quantization"", ""totolook/Quant"", ""FallnAI/Quantize-HF-Models"", ""yasserrmd/MolmoVision"", ""sflindrs/vlm_comparer"", ""ruslanmv/convert_to_gguf"", ""DDDamon/test"", ""quarterturn/molmo-natural-language-image-captioner"", ""awacke1/Leaderboard-Deepseek-Gemini-Grok-GPT-Qwen"", ""sflindrs/Molmo-7B-D-0924-extended-tokens"", ""stoefln/playground2"", ""dkisb/Molmo-7B-D-0924"", ""gaur3009/ColPali-Query-Generator"", ""zainimam/Ss-mol"", ""TheVixhal/OPPE"", ""srawalll/quizzz"", ""vpssud/molmo2"", ""TLPython/MolmoVision"", ""cburtin/ocr"", ""mastercallum/MolmoVision-noCUDAerror"", ""brunocota/Molmo-7B-D-0924"", ""K00B404/LLM_Quantization""], ""safetensors"": {""parameters"": {""F32"": 8021025280}, ""total"": 8021025280}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-04-04 20:01:44+00:00"", ""cardData"": ""base_model:\n- openai/clip-vit-large-patch14-336\n- Qwen/Qwen2-7B\nlanguage:\n- en\nlibrary_name: transformers\nlicense: apache-2.0\npipeline_tag: image-text-to-text\ntags:\n- multimodal\n- olmo\n- molmo\n- pixmo"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": ""modeling_molmo.MolmoForCausalLM"", ""pipeline_tag"": ""text-generation"", ""processor"": null}, ""_id"": ""66f36be6af8609e9ef8198e5"", ""modelId"": ""allenai/Molmo-7B-D-0924"", ""usedStorage"": 35884012544}",0,,0,,0,"https://huggingface.co/cyan2k/molmo-7B-D-bnb-4bit, https://huggingface.co/impactframes/molmo-7B-D-bnb-4bit, https://huggingface.co/detect-tech/molmo-7B-D-bnb-4bit, https://huggingface.co/ctranslate2-4you/molmo-7B-D-0924-bnb-4bit, https://huggingface.co/OPEA/Molmo-7B-D-0924-int4-sym-inc, https://huggingface.co/Scoolar/Molmo-7B-D-0924-NF4",6,,0,"FallnAI/Quantize-HF-Models, K00B404/LLM_Quantization, KBaba7/Quant, TIGER-Lab/MEGA-Bench, akhaliq/Molmo-7B-D-0924, awacke1/Leaderboard-Deepseek-Gemini-Grok-GPT-Qwen, bhaskartripathi/LLM_Quantization, huggingface/InferenceSupport/discussions/new?title=allenai/Molmo-7B-D-0924&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Ballenai%2FMolmo-7B-D-0924%5D(%2Fallenai%2FMolmo-7B-D-0924)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, quarterturn/molmo-natural-language-image-captioner, ruslanmv/convert_to_gguf, sflindrs/vlm_comparer, totolook/Quant, yasserrmd/MolmoVision",13
Nemotron-4-340B-Instruct_finetunes_20250425_125929.csv_finetunes_20250425_125929.csv ADDED
The diff for this file is too large to render. See raw diff
 
Phi-3-mini-128k-instruct_finetunes_20250424_223250.csv_finetunes_20250424_223250.csv ADDED
The diff for this file is too large to render. See raw diff
 
Phi-4-mini-instruct_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv ADDED
The diff for this file is too large to render. See raw diff
 
PixArt-XL-2-1024-MS_finetunes_20250426_215237.csv_finetunes_20250426_215237.csv ADDED
@@ -0,0 +1,329 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ PixArt-alpha/PixArt-XL-2-1024-MS,"---
3
+ license: openrail++
4
+ tags:
5
+ - text-to-image
6
+ - Pixart-α
7
+ ---
8
+
9
+ <p align=""center"">
10
+ <img src=""asset/logo.png"" height=120>
11
+ </p>
12
+
13
+ <div style=""display:flex;justify-content: center"">
14
+ <a href=""https://huggingface.co/spaces/PixArt-alpha/PixArt-alpha""><img src=""https://img.shields.io/static/v1?label=Demo&message=Huggingface&color=yellow""></a> &ensp;
15
+ <a href=""https://pixart-alpha.github.io/""><img src=""https://img.shields.io/static/v1?label=Project%20Page&message=Github&color=blue&logo=github-pages""></a> &ensp;
16
+ <a href=""https://arxiv.org/abs/2310.00426""><img src=""https://img.shields.io/static/v1?label=Paper&message=Arxiv&color=red&logo=arxiv""></a> &ensp;
17
+ <a href=""https://colab.research.google.com/drive/1jZ5UZXk7tcpTfVwnX33dDuefNMcnW9ME?usp=sharing""><img src=""https://img.shields.io/static/v1?label=Free%20Trial&message=Google%20Colab&logo=google&color=orange""></a> &ensp;
18
+ <a href=""https://github.com/orgs/PixArt-alpha/discussions""><img src=""https://img.shields.io/static/v1?label=Discussion&message=Github&color=green&logo=github""></a> &ensp;
19
+ </div>
20
+
21
+ # 🐱 Pixart-α Model Card
22
+ ![row01](asset/images/teaser.png)
23
+
24
+ ## Model
25
+ ![pipeline](asset/images/model.png)
26
+
27
+ [Pixart-α](https://arxiv.org/abs/2310.00426) consists of pure transformer blocks for latent diffusion:
28
+ It can directly generate 1024px images from text prompts within a single sampling process.
29
+
30
+ Source code is available at https://github.com/PixArt-alpha/PixArt-alpha.
31
+
32
+ ### Model Description
33
+
34
+ - **Developed by:** Pixart-α
35
+ - **Model type:** Diffusion-Transformer-based text-to-image generative model
36
+ - **License:** [CreativeML Open RAIL++-M License](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/blob/main/LICENSE.md)
37
+ - **Model Description:** This is a model that can be used to generate and modify images based on text prompts.
38
+ It is a [Transformer Latent Diffusion Model](https://arxiv.org/abs/2310.00426) that uses one fixed, pretrained text encoders ([T5](
39
+ https://huggingface.co/DeepFloyd/t5-v1_1-xxl))
40
+ and one latent feature encoder ([VAE](https://arxiv.org/abs/2112.10752)).
41
+ - **Resources for more information:** Check out our [GitHub Repository](https://github.com/PixArt-alpha/PixArt-alpha) and the [Pixart-α report on arXiv](https://arxiv.org/abs/2310.00426).
42
+
43
+ ### Model Sources
44
+
45
+ For research purposes, we recommend our `generative-models` Github repository (https://github.com/PixArt-alpha/PixArt-alpha),
46
+ which is more suitable for both training and inference and for which most advanced diffusion sampler like [SA-Solver](https://arxiv.org/abs/2309.05019) will be added over time.
47
+ [Hugging Face](https://huggingface.co/spaces/PixArt-alpha/PixArt-alpha) provides free Pixart-α inference.
48
+ - **Repository:** https://github.com/PixArt-alpha/PixArt-alpha
49
+ - **Demo:** https://huggingface.co/spaces/PixArt-alpha/PixArt-alpha
50
+
51
+ # 🔥🔥🔥 Why PixArt-α?
52
+ ## Training Efficiency
53
+ PixArt-α only takes 10.8% of Stable Diffusion v1.5's training time (675 vs. 6,250 A100 GPU days), saving nearly $300,000 ($26,000 vs. $320,000) and reducing 90% CO2 emissions. Moreover, compared with a larger SOTA model, RAPHAEL, our training cost is merely 1%.
54
+ ![Training Efficiency.](asset/images/efficiency.svg)
55
+
56
+ | Method | Type | #Params | #Images | A100 GPU days |
57
+ |-----------|------|---------|---------|---------------|
58
+ | DALL·E | Diff | 12.0B | 1.54B | |
59
+ | GLIDE | Diff | 5.0B | 5.94B | |
60
+ | LDM | Diff | 1.4B | 0.27B | |
61
+ | DALL·E 2 | Diff | 6.5B | 5.63B | 41,66 |
62
+ | SDv1.5 | Diff | 0.9B | 3.16B | 6,250 |
63
+ | GigaGAN | GAN | 0.9B | 0.98B | 4,783 |
64
+ | Imagen | Diff | 3.0B | 15.36B | 7,132 |
65
+ | RAPHAEL | Diff | 3.0B | 5.0B | 60,000 |
66
+ | PixArt-α | Diff | 0.6B | 0.025B | 675 |
67
+
68
+
69
+ ## Evaluation
70
+ ![comparison](asset/images/user-study.png)
71
+ The chart above evaluates user preference for Pixart-α over SDXL 0.9, Stable Diffusion 2, DALLE-2 and DeepFloyd.
72
+ The Pixart-α base model performs comparable or even better than the existing state-of-the-art models.
73
+
74
+
75
+
76
+ ### 🧨 Diffusers
77
+
78
+ Make sure to upgrade diffusers to >= 0.22.0:
79
+ ```
80
+ pip install -U diffusers --upgrade
81
+ ```
82
+
83
+ In addition make sure to install `transformers`, `safetensors`, `sentencepiece`, and `accelerate`:
84
+ ```
85
+ pip install transformers accelerate safetensors sentencepiece
86
+ ```
87
+
88
+ To just use the base model, you can run:
89
+
90
+
91
+ ```py
92
+ from diffusers import PixArtAlphaPipeline
93
+ import torch
94
+
95
+ pipe = PixArtAlphaPipeline.from_pretrained(""PixArt-alpha/PixArt-XL-2-1024-MS"", torch_dtype=torch.float16)
96
+ pipe = pipe.to(""cuda"")
97
+
98
+ # if using torch < 2.0
99
+ # pipe.enable_xformers_memory_efficient_attention()
100
+
101
+ prompt = ""An astronaut riding a green horse""
102
+ images = pipe(prompt=prompt).images[0]
103
+ ```
104
+
105
+ When using `torch >= 2.0`, you can improve the inference speed by 20-30% with torch.compile. Simple wrap the unet with torch compile before running the pipeline:
106
+ ```py
107
+ pipe.transformer = torch.compile(pipe.transformer, mode=""reduce-overhead"", fullgraph=True)
108
+ ```
109
+
110
+ If you are limited by GPU VRAM, you can enable *cpu offloading* by calling `pipe.enable_model_cpu_offload`
111
+ instead of `.to(""cuda"")`:
112
+
113
+ ```diff
114
+ - pipe.to(""cuda"")
115
+ + pipe.enable_model_cpu_offload()
116
+ ```
117
+
118
+ For more information on how to use Pixart-α with `diffusers`, please have a look at [the Pixart-α Docs](https://huggingface.co/docs/diffusers/main/en/api/pipelines/pixart).
119
+
120
+ ### Free Google Colab
121
+ You can use Google Colab to generate images from PixArt-α free of charge. Click [here](https://colab.research.google.com/drive/1jZ5UZXk7tcpTfVwnX33dDuefNMcnW9ME?usp=sharing) to try.
122
+
123
+ ## Uses
124
+
125
+ ### Direct Use
126
+
127
+ The model is intended for research purposes only. Possible research areas and tasks include
128
+
129
+ - Generation of artworks and use in design and other artistic processes.
130
+ - Applications in educational or creative tools.
131
+ - Research on generative models.
132
+ - Safe deployment of models which have the potential to generate harmful content.
133
+
134
+ - Probing and understanding the limitations and biases of generative models.
135
+
136
+ Excluded uses are described below.
137
+
138
+ ### Out-of-Scope Use
139
+
140
+ The model was not trained to be factual or true representations of people or events, and therefore using the model to generate such content is out-of-scope for the abilities of this model.
141
+
142
+ ## Limitations and Bias
143
+
144
+ ### Limitations
145
+
146
+
147
+ - The model does not achieve perfect photorealism
148
+ - The model cannot render legible text
149
+ - The model struggles with more difficult tasks which involve compositionality, such as rendering an image corresponding to “A red cube on top of a blue sphere”
150
+ - fingers, .etc in general may not be generated properly.
151
+ - The autoencoding part of the model is lossy.
152
+
153
+ ### Bias
154
+ While the capabilities of image generation models are impressive, they can also reinforce or exacerbate social biases.
155
+ ","{""id"": ""PixArt-alpha/PixArt-XL-2-1024-MS"", ""author"": ""PixArt-alpha"", ""sha"": ""b89adadeccd9ead2adcb9fa2825d3fabec48d404"", ""last_modified"": ""2023-11-07 06:11:50+00:00"", ""created_at"": ""2023-11-04 15:48:30+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 122196, ""downloads_all_time"": null, ""likes"": 201, ""library_name"": ""diffusers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""diffusers"", ""safetensors"", ""text-to-image"", ""Pixart-\u03b1"", ""arxiv:2310.00426"", ""arxiv:2112.10752"", ""arxiv:2309.05019"", ""license:openrail++"", ""endpoints_compatible"", ""diffusers:PixArtAlphaPipeline"", ""region:us""], ""pipeline_tag"": ""text-to-image"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""license: openrail++\ntags:\n- text-to-image\n- Pixart-\u03b1"", ""widget_data"": null, ""model_index"": null, ""config"": {""diffusers"": {""_class_name"": ""PixArtAlphaPipeline""}}, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/PixArt.svg', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/examples.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/images/controlnet/controlnet_huawei.svg', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/images/controlnet/controlnet_iclr.svg', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/images/controlnet/controlnet_lenna.svg', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/images/dreambooth/dreambooth_dog.svg', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/images/dreambooth/dreambooth_m5.svg', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/images/efficiency.svg', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/images/model.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/images/more-samples.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/images/more-samples1.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/images/sample.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/images/teaser.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/images/user-study.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/logo.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='asset/samples.txt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model_index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='scheduler/scheduler_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='text_encoder/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='text_encoder/model-00001-of-00002.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='text_encoder/model-00002-of-00002.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='text_encoder/model.safetensors.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer/added_tokens.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer/special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer/spiece.model', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer/tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='transformer/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='transformer/diffusion_pytorch_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='vae/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='vae/diffusion_pytorch_model.safetensors', size=None, blob_id=None, lfs=None)""], ""spaces"": [""Lightricks/LTX-Video-Playground"", ""PixArt-alpha/PixArt-alpha"", ""TIGER-Lab/GenAI-Arena"", ""artificialguybr/Pixart-Sigma"", ""alibaba-pai/EasyAnimate"", ""LanguageBind/Open-Sora-Plan-v1.0.0"", ""TencentARC/ColorFlow"", ""jasperai/flash-diffusion"", ""fantos/VoiceClone"", ""LanguageBind/Open-Sora-Plan-v1.1.0"", ""Nymbo/image_gen_supaqueue"", ""fantaxy/fastvideogen"", ""fffiloni/ReNO"", ""maxin-cn/Latte-1"", ""fffiloni/flash-wallpapers"", ""cocktailpeanut/LTX-Video-Playground"", ""diffusers/compute-pipeline-size"", ""ali-vilab/IDEA-Bench-Arena"", ""priyanshu9588/PixArt-alpha"", ""ford442/LTX-Video"", ""1inkusFace/LTX-Video-Xora"", ""openfree/ginigen-sora"", ""svjack/LTX-Video-Playground"", ""akthangdz/TEXT_TO_VIDEO"", ""JunhaoZhuang/Cobra"", ""sidd-genmo/Open-Sora-Plan-v1.0.0"", ""jarnot/EasyAnimate"", ""PeepDaSlan9/HYDRAS_Latte-1"", ""emilalvaro/LTX-Video-Playground"", ""Kaskatraz/LTX-Video-Playground"", ""jalve/jalvneis"", ""dd890/PixArt-alpha-PixArt-XL-2-1024-MS"", ""jalve/NeisAlv"", ""MrOvkill/PixArt-alpha-moddedalltohell"", ""YanzBotz/PixArt"", ""wandb/reproducible-pixart-alpha"", ""vakilrathod67/PixArt-alpha-PixArt-XL-2-1024-MS"", ""Jyothirmai782/Pixart-Sigma"", ""Viswanath999/Pixart-Sigma"", ""Taf2023/Open-Sora-Plan-v1.0.0"", ""lylosn/Open-Sora-Plan-v1.0.0"", ""tsi-org/PixArt-alpha"", ""lcyyyy/homework_end"", ""tsi-org/PixioArt-alpha"", ""CPM1234567890/ex01"", ""yufiofficial/PixArt-alpha-PixArt-XL-2-1024-MS"", ""Lucas94/PixArt-alpha-PixArt-XL-2-1024-MS"", ""RO-Rtechs/Rtechs_Open-Sora-Plan-v1.1.0"", ""cocktailpeanut/flash-diffusion"", ""BobLLM/Sora"", ""kletoskletos/PixArt-alpha-PixArt-XL-2-1024-MS"", ""Nymbo/flash-wallpapers"", ""Dragunflie-420/flash-diffusion"", ""K00B404/image_gen_supaqueue_game_assets"", ""YuwanA55/Flash_Jasper"", ""K00B404/EasyAnimate_custom"", ""pang1368/PixArt-alpha-PixArt-XL-2-1024-MS"", ""jbilcke-hf/ai-tube-model-ltxv-1"", ""NativeAngels/LTX-Video-Playground"", ""Swaqgame99/LTX-Video-Playground"", ""jbilcke-hf/ai-tube-model-ltxv-2"", ""jbilcke-hf/ai-tube-model-ltxv-3"", ""k11112/LTX-Video-Playground"", ""jbilcke-hf/ai-tube-model-ltxv-4"", ""AashishNKumar/proj11"", ""riflecreek/LTX-Video-Playground"", ""kostadinkostad/LTX-Video-Playground"", ""K00B404/LTX-Video-Playground"", ""constant999/LTX-Video-Playground"", ""svjack/ColorFlow"", ""Lezzio/LTX-Video-Playground"", ""Greekmongoose/LTX-Video-Playground"", ""PeldelnasGranell/LTX-Video-Playground"", ""learningloop/LTX-Video-Playground"", ""bestoai/PixArt-alpha"", ""Kremon96/VoiceClone"", ""waloneai/fastvideogen"", ""Tusharcrusty1/LTX-Video-Playground"", ""sitonmyface/LTX-Video-Playground"", ""nasiye969/LTX-Video-Playground"", ""Jwrockon/ArtemisVoiceCloner"", ""WillybotAI77777/LTX-Video-Playground"", ""peterpeter8585/VoiceClone"", ""MrDrmm/EasyAnimate"", ""waloneai/EasyAnimate"", ""Pablosolaris/LTX-Video-Playground"", ""stepbysteb/EasyAnimate"", ""cziter15pl/LTX-Video-Playground"", ""kahramango/EasyAnimate"", ""daaaaaaaaaaa2/LTX-Video-Playground"", ""data97688/PixArt-alpha-PixArt-XL-2-1024-MS"", ""jewelt123/VoiceCloneABC2"", ""vyloup/LTX-Video-Playground"", ""dotkaio/LTX-Video-Playground"", ""Quantamhash/Quantam_Clone"", ""dalouniquefans/dalo-image-api"", ""Rakeshj182002/LTX-Video-Playground""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-11-07 06:11:50+00:00"", ""cardData"": ""license: openrail++\ntags:\n- text-to-image\n- Pixart-\u03b1"", ""transformersInfo"": null, ""_id"": ""654667ce9c4bf757d6d2ff63"", ""modelId"": ""PixArt-alpha/PixArt-XL-2-1024-MS"", ""usedStorage"": 24293837816}",0,"https://huggingface.co/JunhaoZhuang/Cobra, https://huggingface.co/Luo-Yihong/yoso_pixart1024",2,https://huggingface.co/jasperai/flash-pixart,1,https://huggingface.co/calcuis/pixart,1,,0,"Lightricks/LTX-Video-Playground, Nymbo/image_gen_supaqueue, PixArt-alpha/PixArt-alpha, TIGER-Lab/GenAI-Arena, TencentARC/ColorFlow, ali-vilab/IDEA-Bench-Arena, alibaba-pai/EasyAnimate, artificialguybr/Pixart-Sigma, diffusers/compute-pipeline-size, fantaxy/fastvideogen, fantos/VoiceClone, huggingface/InferenceSupport/discussions/new?title=PixArt-alpha/PixArt-XL-2-1024-MS&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BPixArt-alpha%2FPixArt-XL-2-1024-MS%5D(%2FPixArt-alpha%2FPixArt-XL-2-1024-MS)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, maxin-cn/Latte-1, openfree/ginigen-sora",14
156
+ JunhaoZhuang/Cobra,"---
157
+ base_model:
158
+ - PixArt-alpha/PixArt-XL-2-1024-MS
159
+ language:
160
+ - en
161
+ license: apache-2.0
162
+ pipeline_tag: image-to-image
163
+ library_name: diffusers
164
+ ---
165
+
166
+ # 🎨 Cobra
167
+
168
+ **Efficient Line Art COlorization with BRoAder References**
169
+
170
+ **Authors:** Junhao Zhuang, Lingen Li, Xuan Ju, Zhaoyang Zhang, Chun Yuan† and Ying Shan†
171
+
172
+ <a href='https://zhuang2002.github.io/Cobra/'><img src='https://img.shields.io/badge/Project-Page-Green'></a> &nbsp;
173
+ <a href=""https://github.com/Zhuang2002/Cobra""><img src=""https://img.shields.io/badge/GitHub-Repository-black?logo=github""></a> &nbsp;
174
+ <a href='https://huggingface.co/spaces/JunhaoZhuang/Cobra'><img src='https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Demo-blue'></a> &nbsp;
175
+ <a href=""https://arxiv.org/abs/2504.12240""><img src=""https://img.shields.io/badge/arXiv-2504.12240-b31b1b.svg""></a> &nbsp;
176
+ <a href=""https://huggingface.co/JunhaoZhuang/Cobra""><img src=""https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Model-blue""></a>
177
+
178
+ **Your star means a lot for us to develop this project!** :star:
179
+
180
+ <img src='https://zhuang2002.github.io/Cobra/fig/teaser.png'/>
181
+
182
+ ### 🌟 Abstract
183
+
184
+ The comic production industry requires reference-based line art colorization with high accuracy, efficiency, contextual consistency, and flexible control.
185
+ A comic page often involves diverse characters, objects, and backgrounds, which complicates the coloring process.
186
+ Despite advancements in diffusion models for image generation, their application in line art colorization remains limited, facing challenges related to handling extensive reference images, time-consuming inference, and flexible control.
187
+ We investigate the necessity of extensive contextual image guidance on the quality of line art colorization. To address these challenges, we introduce **Cobra**, an efficient and versatile method that supports color hints and utilizes **over 200 reference images** while maintaining low latency.
188
+ Central to Cobra is a Causal Sparse DiT architecture, which leverages specially designed positional encodings, causal sparse attention, and Key-Value Cache to effectively manage long-context references and ensure color identity consistency.
189
+ Results demonstrate that Cobra achieves accurate line art colorization through extensive contextual reference, significantly enhancing inference speed and interactivity, thereby meeting critical industrial demands.
190
+
191
+ ### 📰 News
192
+
193
+ - **Release Date:** April 17, 2025 - The inference code and model weights have also been released! 🎉
194
+
195
+ ### 📋 TODO
196
+
197
+ - ✅ Release inference code and model weights
198
+ - ⬜️ Release training code
199
+
200
+ ### 🚀 Getting Started
201
+
202
+ Follow these steps to set up and run Cobra on your local machine:
203
+
204
+ - **Clone the Repository**
205
+
206
+ Download the code from our GitHub repository:
207
+ ```bash
208
+ git clone https://github.com/zhuang2002/Cobra
209
+ cd Cobra
210
+ ```
211
+
212
+ - **Set Up the Python Environment**
213
+
214
+ Ensure you have Anaconda or Miniconda installed, then create and activate a Python environment and install required dependencies:
215
+ ```bash
216
+ conda create -n cobra python=3.11.11
217
+ conda activate cobra
218
+ pip install -r requirements.txt
219
+ ```
220
+
221
+ - **Run the Application**
222
+
223
+ You can launch the Gradio interface for Cobra by running the following command:
224
+ ```bash
225
+ python app.py
226
+ ```
227
+
228
+ - **Access Cobra in Your Browser**
229
+
230
+ Open your browser and go to `http://localhost:7860`. If you're running the app on a remote server, replace `localhost` with your server's IP address or domain name. To use a custom port, update the `server_port` parameter in the `demo.launch()` function of app.py.
231
+
232
+ ### 🎉 Demo
233
+
234
+ You can [try the demo](https://huggingface.co/spaces/JunhaoZhuang/Cobra) of Cobra on Hugging Face Space.
235
+
236
+ ### 🛠️ Method
237
+
238
+ The overview of Cobra.
239
+ This figure depicts the framework of Cobra, which utilizes a large collection of retrieved reference images to guide the colorization of comic line art. The framework effectively manages an arbitrary number of contextual image references through localized reusable positional encoding, ensuring appropriate aspect ratios and resolutions. Additionally, the causal sparse DiT architecture processes long contextual references, enhancing identity preservation and color accuracy while reducing computational complexity. The integration of optional color hints further ensures user flexibility, culminating in high-quality coloring that is highly suitable for industrial applications.
240
+
241
+ <img src=""https://zhuang2002.github.io/Cobra/fig/flowchart.png"" width=""1000"">
242
+
243
+ 🤗 We welcome your feedback, questions, or collaboration opportunities. Thank you for trying Cobra!
244
+
245
+ ### 📄 Acknowledgments
246
+
247
+ We would like to acknowledge the following open-source projects that have inspired and contributed to the development of Cobra:
248
+
249
+ - **MangaLineExtraction_PyTorch**: https://github.com/ljsabc/MangaLineExtraction_PyTorch
250
+
251
+ We are grateful for the valuable resources and insights provided by these projects.
252
+
253
+ ### 📞 Contact
254
+
255
+ - **Junhao Zhuang**
256
+ Email: [zhuangjh23@mails.tsinghua.edu.cn](mailto:zhuangjh23@mails.tsinghua.edu.cn)
257
+
258
+ ### 📜 Citation
259
+
260
+ ```
261
+ @misc{zhuang2025cobraefficientlineart,
262
+ title={Cobra: Efficient Line Art COlorization with BRoAder References},
263
+ author={Junhao Zhuang and Lingen Li and Xuan Ju and Zhaoyang Zhang and Chun Yuan and Ying Shan},
264
+ year={2025},
265
+ eprint={2504.12240},
266
+ archivePrefix={arXiv},
267
+ primaryClass={cs.CV},
268
+ url={https://arxiv.org/abs/2504.12240},
269
+ }
270
+ ```","{""id"": ""JunhaoZhuang/Cobra"", ""author"": ""JunhaoZhuang"", ""sha"": ""8451af08de9224649b63aa50980fd5f9ac30d57e"", ""last_modified"": ""2025-04-17 12:47:24+00:00"", ""created_at"": ""2025-04-12 07:46:07+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 16, ""library_name"": ""diffusers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""diffusers"", ""image-to-image"", ""en"", ""arxiv:2504.12240"", ""base_model:PixArt-alpha/PixArt-XL-2-1024-MS"", ""base_model:finetune:PixArt-alpha/PixArt-XL-2-1024-MS"", ""license:apache-2.0"", ""region:us""], ""pipeline_tag"": ""image-to-image"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- PixArt-alpha/PixArt-XL-2-1024-MS\nlanguage:\n- en\nlibrary_name: diffusers\nlicense: apache-2.0\npipeline_tag: image-to-image"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='LE/erika.pth', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image_encoder/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image_encoder/pytorch_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='line_GSRP/MultiResNetModel.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='line_ckpt/controlnet.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='line_ckpt/transformer_lora_pos.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='shadow_GSRP/MultiResNetModel.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='shadow_ckpt/controlnet.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='shadow_ckpt/transformer_lora_pos.bin', size=None, blob_id=None, lfs=None)""], ""spaces"": [""JunhaoZhuang/Cobra""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-04-17 12:47:24+00:00"", ""cardData"": ""base_model:\n- PixArt-alpha/PixArt-XL-2-1024-MS\nlanguage:\n- en\nlibrary_name: diffusers\nlicense: apache-2.0\npipeline_tag: image-to-image"", ""transformersInfo"": null, ""_id"": ""67fa1a3f59fef5be4e929c75"", ""modelId"": ""JunhaoZhuang/Cobra"", ""usedStorage"": 4831835390}",1,,0,,0,,0,,0,"JunhaoZhuang/Cobra, huggingface/InferenceSupport/discussions/new?title=JunhaoZhuang/Cobra&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BJunhaoZhuang%2FCobra%5D(%2FJunhaoZhuang%2FCobra)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A",2
271
+ Luo-Yihong/yoso_pixart1024,"---
272
+ language:
273
+ - en
274
+ library_name: diffusers
275
+ pipeline_tag: text-to-image
276
+ base_model:
277
+ - PixArt-alpha/PixArt-XL-2-1024-MS
278
+ ---
279
+ # You Only Sample Once (YOSO)
280
+
281
+ ![overview](overview.jpg)
282
+
283
+ The YOSO was proposed in ""[You Only Sample Once: Taming One-Step Text-To-Image Synthesis by Self-Cooperative Diffusion GANs](https://www.arxiv.org/abs/2403.12931)"" by *Yihong Luo, Xiaolong Chen, Xinghua Qu, Jing Tang*.
284
+
285
+ Official Repository of this paper: [YOSO](https://github.com/Luo-Yihong/YOSO).
286
+
287
+ This model is fine-tuning from [
288
+ PixArt-XL-2-512x512](https://huggingface.co/PixArt-alpha/PixArt-XL-2-512x512), enabling one-step inference to perform text-to-image generation.
289
+
290
+ We wanna highlight that the YOSO-PixArt was originally trained on 512 resolution. However, we found that we can construct a YOSO that enables generating samples with 1024 resolution by merging with [
291
+ PixArt-XL-2-1024-MS](https://huggingface.co/PixArt-alpha/PixArt-XL-2-1024-MS
292
+ ) (Section 6.3.1 in the paper). The impressive performance indicates the robust generalization ability of our YOSO.
293
+ ## usage
294
+ ```python
295
+ import torch
296
+ from diffusers import PixArtAlphaPipeline, LCMScheduler, Transformer2DModel
297
+
298
+ transformer = Transformer2DModel.from_pretrained(
299
+ ""Luo-Yihong/yoso_pixart1024"", torch_dtype=torch.float16).to('cuda')
300
+
301
+ pipe = PixArtAlphaPipeline.from_pretrained(""PixArt-alpha/PixArt-XL-2-512x512"",
302
+ transformer=transformer,
303
+ torch_dtype=torch.float16, use_safetensors=True)
304
+
305
+ pipe = pipe.to('cuda')
306
+ pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config)
307
+ pipe.scheduler.config.prediction_type = ""v_prediction""
308
+ generator = torch.manual_seed(318)
309
+ imgs = pipe(prompt=""Pirate ship trapped in a cosmic maelstrom nebula, rendered in cosmic beach whirlpool engine, volumetric lighting, spectacular, ambient lights, light pollution, cinematic atmosphere, art nouveau style, illustration art artwork by SenseiJaye, intricate detail."",
310
+ num_inference_steps=1,
311
+ num_images_per_prompt = 1,
312
+ generator = generator,
313
+ guidance_scale=1.,
314
+ )[0]
315
+ imgs[0]
316
+ ```
317
+ ![Ship](ship_1024.jpg)
318
+
319
+ ## Bibtex
320
+ ```
321
+ @misc{luo2024sample,
322
+ title={You Only Sample Once: Taming One-Step Text-to-Image Synthesis by Self-Cooperative Diffusion GANs},
323
+ author={Yihong Luo and Xiaolong Chen and Xinghua Qu and Jing Tang},
324
+ year={2024},
325
+ eprint={2403.12931},
326
+ archivePrefix={arXiv},
327
+ primaryClass={cs.CV}
328
+ }
329
+ ```","{""id"": ""Luo-Yihong/yoso_pixart1024"", ""author"": ""Luo-Yihong"", ""sha"": ""f2cadc6f1c81c30f7b82feca1442fc0601dbdcaa"", ""last_modified"": ""2025-03-16 17:01:02+00:00"", ""created_at"": ""2024-03-18 08:23:34+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 19, ""downloads_all_time"": null, ""likes"": 12, ""library_name"": ""diffusers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""diffusers"", ""safetensors"", ""text-to-image"", ""en"", ""arxiv:2403.12931"", ""base_model:PixArt-alpha/PixArt-XL-2-1024-MS"", ""base_model:finetune:PixArt-alpha/PixArt-XL-2-1024-MS"", ""region:us""], ""pipeline_tag"": ""text-to-image"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- PixArt-alpha/PixArt-XL-2-1024-MS\nlanguage:\n- en\nlibrary_name: diffusers\npipeline_tag: text-to-image"", ""widget_data"": null, ""model_index"": null, ""config"": {}, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='diffusion_pytorch_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='overview.jpg', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ship_1024.jpg', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-03-16 17:01:02+00:00"", ""cardData"": ""base_model:\n- PixArt-alpha/PixArt-XL-2-1024-MS\nlanguage:\n- en\nlibrary_name: diffusers\npipeline_tag: text-to-image"", ""transformersInfo"": null, ""_id"": ""65f7fa06dd3cc437a8a30e74"", ""modelId"": ""Luo-Yihong/yoso_pixart1024"", ""usedStorage"": 2453869313}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=Luo-Yihong/yoso_pixart1024&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BLuo-Yihong%2Fyoso_pixart1024%5D(%2FLuo-Yihong%2Fyoso_pixart1024)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
SD_PixelArt_SpriteSheet_Generator_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ Onodofthenorth/SD_PixelArt_SpriteSheet_Generator,"---
3
+ license: apache-2.0
4
+ language:
5
+ - en
6
+ pipeline_tag: text-to-image
7
+ tags:
8
+ - spritesheet
9
+ - text-to-image
10
+ ---
11
+ This Stable diffusion checkpoint allows you to generate pixel art sprite sheets from four different angles.
12
+ These first images are my results after merging this model with another model trained on my wife. merging another model with this one is the easiest way to get a consistent character with each view. still requires a bit of playing around with settings in img2img to get them how you want. for left and right, I suggest picking your best result and mirroring. after you are satisfied take your photo into photoshop or Krita, remove the background, and scale to the desired size. after this you can scale back up to display your results; this also clears up some of the color murkiness in the initial outputs.
13
+ ![final sheet.png](https://s3.amazonaws.com/moonup/production/uploads/1667278292305-63028bc42db53f7d9f38dadb.png)
14
+
15
+ ### 🧨 Diffusers
16
+
17
+ This model can be used just like any other Stable Diffusion model. For more information,
18
+ please have a look at the [Stable Diffusion](https://huggingface.co/docs/diffusers/api/pipelines/stable_diffusion).
19
+
20
+ You can also export the model to [ONNX](https://huggingface.co/docs/diffusers/optimization/onnx), [MPS](https://huggingface.co/docs/diffusers/optimization/mps) and/or [FLAX/JAX]().
21
+
22
+ ```python
23
+ #!pip install diffusers transformers scipy torch
24
+ from diffusers import StableDiffusionPipeline
25
+ import torch
26
+ model_id = ""Onodofthenorth/SD_PixelArt_SpriteSheet_Generator""
27
+ pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16)
28
+ pipe = pipe.to(""cuda"")
29
+ prompt = ""PixelartLSS""
30
+ image = pipe(prompt).images[0]
31
+ image.save(""./pixel.png"")
32
+ ```
33
+ ___
34
+ ___
35
+ For the front view use ""PixelartFSS""
36
+ ![Layer-0.png](https://s3.amazonaws.com/moonup/production/uploads/1667278301151-63028bc42db53f7d9f38dadb.png)
37
+ ___
38
+ ___
39
+ For the right view use ""PixelartRSS""
40
+ ![tmpjn2e6lle-001.png](https://s3.amazonaws.com/moonup/production/uploads/1667278489924-63028bc42db53f7d9f38dadb.png)
41
+ ___
42
+ ___
43
+ For the back view use ""PixelartBSS""
44
+ ![tmpxowqvkbj-001.png](https://s3.amazonaws.com/moonup/production/uploads/1667278497366-63028bc42db53f7d9f38dadb.png)
45
+ ___
46
+ ___
47
+ For the left view use ""PixelartLSS""
48
+ ![Left-001.png](https://s3.amazonaws.com/moonup/production/uploads/1667278508076-63028bc42db53f7d9f38dadb.png)
49
+ ___
50
+ ___
51
+ These are random results from the unmerged model
52
+ ![pixel art full.png](https://s3.amazonaws.com/moonup/production/uploads/1667278579019-63028bc42db53f7d9f38dadb.png)
53
+ ___
54
+ ___
55
+ here's a result from a merge with my Hermione model
56
+ ![00711-1613350423-PixelartFSS, hermione.png](https://s3.amazonaws.com/moonup/production/uploads/1667278700840-63028bc42db53f7d9f38dadb.png)
57
+ ___
58
+ ___
59
+ here's a result from a merge with my cat girl model
60
+ ![00675-557755173-(catgirl) PixelartLSS.png](https://s3.amazonaws.com/moonup/production/uploads/1667278670012-63028bc42db53f7d9f38dadb.png)","{""id"": ""Onodofthenorth/SD_PixelArt_SpriteSheet_Generator"", ""author"": ""Onodofthenorth"", ""sha"": ""8229c9b6e928103f0e657cfe6b14d902cb2101d6"", ""last_modified"": ""2023-05-05 18:30:10+00:00"", ""created_at"": ""2022-11-01 04:31:21+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 1501, ""downloads_all_time"": null, ""likes"": 429, ""library_name"": ""diffusers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""diffusers"", ""spritesheet"", ""text-to-image"", ""en"", ""license:apache-2.0"", ""autotrain_compatible"", ""endpoints_compatible"", ""diffusers:StableDiffusionPipeline"", ""region:us""], ""pipeline_tag"": ""text-to-image"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""language:\n- en\nlicense: apache-2.0\npipeline_tag: text-to-image\ntags:\n- spritesheet\n- text-to-image"", ""widget_data"": null, ""model_index"": null, ""config"": {""diffusers"": {""_class_name"": ""StableDiffusionPipeline""}}, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='PixelartSpritesheet_V.1.ckpt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='feature_extractor/preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model_index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='safety_checker/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='safety_checker/pytorch_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='scheduler/scheduler_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='text_encoder/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='text_encoder/pytorch_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer/merges.txt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer/special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer/tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer/vocab.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='unet/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='unet/diffusion_pytorch_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='vae/config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='vae/diffusion_pytorch_model.bin', size=None, blob_id=None, lfs=None)""], ""spaces"": [""ronvolutional/sd-spritesheets"", ""Nymbo/image_gen_supaqueue"", ""Samuelblue/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator"", ""CharlieAmalet/Tools3ox_PixelArt_SpriteSheet_GeneratorArt_Api"", ""DEVINKofficial/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator"", ""Derni/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator"", ""tomhitto/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator"", ""bspSHU/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator"", ""sanchezNa/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator"", ""studentofplato/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator"", ""Bready11/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator"", ""Ryanforbus/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator"", ""johnson1984/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator"", ""ASsazz/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator"", ""Lubub/sd-spritesheets"", ""LububMalvino/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator"", ""CLOUDWERXLAB/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator"", ""K00B404/image_gen_supaqueue_game_assets"", ""SrKatayama/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-05-05 18:30:10+00:00"", ""cardData"": ""language:\n- en\nlicense: apache-2.0\npipeline_tag: text-to-image\ntags:\n- spritesheet\n- text-to-image"", ""transformersInfo"": null, ""_id"": ""6360a119773df6f83db24210"", ""modelId"": ""Onodofthenorth/SD_PixelArt_SpriteSheet_Generator"", ""usedStorage"": 19492912354}",0,,0,,0,,0,,0,"Bready11/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator, CLOUDWERXLAB/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator, CharlieAmalet/Tools3ox_PixelArt_SpriteSheet_GeneratorArt_Api, DEVINKofficial/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator, Derni/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator, Nymbo/image_gen_supaqueue, Samuelblue/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator, bspSHU/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator, huggingface/InferenceSupport/discussions/new?title=Onodofthenorth/SD_PixelArt_SpriteSheet_Generator&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BOnodofthenorth%2FSD_PixelArt_SpriteSheet_Generator%5D(%2FOnodofthenorth%2FSD_PixelArt_SpriteSheet_Generator)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, ronvolutional/sd-spritesheets, sanchezNa/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator, studentofplato/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator, tomhitto/Onodofthenorth-SD_PixelArt_SpriteSheet_Generator",13
SFR-Embedding-Mistral_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv ADDED
The diff for this file is too large to render. See raw diff
 
SSD-1B_finetunes_20250424_223250.csv_finetunes_20250424_223250.csv ADDED
The diff for this file is too large to render. See raw diff
 
SenseVoiceSmall_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv ADDED
@@ -0,0 +1,235 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ FunAudioLLM/SenseVoiceSmall,"---
3
+ license: other
4
+ license_name: model-license
5
+ license_link: https://github.com/modelscope/FunASR/blob/main/MODEL_LICENSE
6
+ language:
7
+ - en
8
+ - zh
9
+ - ja
10
+ - ko
11
+ library: funasr
12
+ ---
13
+ ([简体中文](./README_zh.md)|English|[日本語](./README_ja.md))
14
+
15
+
16
+ # Introduction
17
+
18
+ github [repo](https://github.com/FunAudioLLM/SenseVoice) : https://github.com/FunAudioLLM/SenseVoice
19
+
20
+ SenseVoice is a speech foundation model with multiple speech understanding capabilities, including automatic speech recognition (ASR), spoken language identification (LID), speech emotion recognition (SER), and audio event detection (AED).
21
+
22
+ <img src=""image/sensevoice2.png"">
23
+
24
+ [//]: # (<div align=""center""><img src=""image/sensevoice.png"" width=""700""/> </div>)
25
+
26
+ <div align=""center"">
27
+ <h4>
28
+ <a href=""https://fun-audio-llm.github.io/""> Homepage </a>
29
+ ｜<a href=""#What's News""> What's News </a>
30
+ ｜<a href=""#Benchmarks""> Benchmarks </a>
31
+ ｜<a href=""#Install""> Install </a>
32
+ ｜<a href=""#Usage""> Usage </a>
33
+ ｜<a href=""#Community""> Community </a>
34
+ </h4>
35
+
36
+ Model Zoo:
37
+ [modelscope](https://www.modelscope.cn/models/iic/SenseVoiceSmall), [huggingface](https://huggingface.co/FunAudioLLM/SenseVoiceSmall)
38
+
39
+ Online Demo:
40
+ [modelscope demo](https://www.modelscope.cn/studios/iic/SenseVoice), [huggingface space](https://huggingface.co/spaces/FunAudioLLM/SenseVoice)
41
+
42
+
43
+ </div>
44
+
45
+
46
+ <a name=""Highligts""></a>
47
+ # Highlights 🎯
48
+ **SenseVoice** focuses on high-accuracy multilingual speech recognition, speech emotion recognition, and audio event detection.
49
+ - **Multilingual Speech Recognition:** Trained with over 400,000 hours of data, supporting more than 50 languages, the recognition performance surpasses that of the Whisper model.
50
+ - **Rich transcribe:**
51
+ - Possess excellent emotion recognition capabilities, achieving and surpassing the effectiveness of the current best emotion recognition models on test data.
52
+ - Offer sound event detection capabilities, supporting the detection of various common human-computer interaction events such as bgm, applause, laughter, crying, coughing, and sneezing.
53
+ - **Efficient Inference:** The SenseVoice-Small model utilizes a non-autoregressive end-to-end framework, leading to exceptionally low inference latency. It requires only 70ms to process 10 seconds of audio, which is 15 times faster than Whisper-Large.
54
+ - **Convenient Finetuning:** Provide convenient finetuning scripts and strategies, allowing users to easily address long-tail sample issues according to their business scenarios.
55
+ - **Service Deployment:** Offer service deployment pipeline, supporting multi-concurrent requests, with client-side languages including Python, C++, HTML, Java, and C#, among others.
56
+
57
+ <a name=""What's News""></a>
58
+ # What's New 🔥
59
+ - 2024/7: Added Export Features for [ONNX](https://github.com/FunAudioLLM/SenseVoice/demo_onnx.py) and [libtorch](https://github.com/FunAudioLLM/SenseVoice/demo_libtorch.py), as well as Python Version Runtimes: [funasr-onnx-0.4.0](https://pypi.org/project/funasr-onnx/), [funasr-torch-0.1.1](https://pypi.org/project/funasr-torch/)
60
+ - 2024/7: The [SenseVoice-Small](https://www.modelscope.cn/models/iic/SenseVoiceSmall) voice understanding model is open-sourced, which offers high-precision multilingual speech recognition, emotion recognition, and audio event detection capabilities for Mandarin, Cantonese, English, Japanese, and Korean and leads to exceptionally low inference latency.
61
+ - 2024/7: The CosyVoice for natural speech generation with multi-language, timbre, and emotion control. CosyVoice excels in multi-lingual voice generation, zero-shot voice generation, cross-lingual voice cloning, and instruction-following capabilities. [CosyVoice repo](https://github.com/FunAudioLLM/CosyVoice) and [CosyVoice space](https://www.modelscope.cn/studios/iic/CosyVoice-300M).
62
+ - 2024/7: [FunASR](https://github.com/modelscope/FunASR) is a fundamental speech recognition toolkit that offers a variety of features, including speech recognition (ASR), Voice Activity Detection (VAD), Punctuation Restoration, Language Models, Speaker Verification, Speaker Diarization and multi-talker ASR.
63
+
64
+ <a name=""Benchmarks""></a>
65
+ # Benchmarks 📝
66
+
67
+ ## Multilingual Speech Recognition
68
+ We compared the performance of multilingual speech recognition between SenseVoice and Whisper on open-source benchmark datasets, including AISHELL-1, AISHELL-2, Wenetspeech, LibriSpeech, and Common Voice. In terms of Chinese and Cantonese recognition, the SenseVoice-Small model has advantages.
69
+
70
+ <div align=""center"">
71
+ <img src=""image/asr_results1.png"" width=""400"" /><img src=""image/asr_results2.png"" width=""400"" />
72
+ </div>
73
+
74
+ ## Speech Emotion Recognition
75
+
76
+ Due to the current lack of widely-used benchmarks and methods for speech emotion recognition, we conducted evaluations across various metrics on multiple test sets and performed a comprehensive comparison with numerous results from recent benchmarks. The selected test sets encompass data in both Chinese and English, and include multiple styles such as performances, films, and natural conversations. Without finetuning on the target data, SenseVoice was able to achieve and exceed the performance of the current best speech emotion recognition models.
77
+
78
+ <div align=""center"">
79
+ <img src=""image/ser_table.png"" width=""1000"" />
80
+ </div>
81
+
82
+ Furthermore, we compared multiple open-source speech emotion recognition models on the test sets, and the results indicate that the SenseVoice-Large model achieved the best performance on nearly all datasets, while the SenseVoice-Small model also surpassed other open-source models on the majority of the datasets.
83
+
84
+ <div align=""center"">
85
+ <img src=""image/ser_figure.png"" width=""500"" />
86
+ </div>
87
+
88
+ ## Audio Event Detection
89
+
90
+ Although trained exclusively on speech data, SenseVoice can still function as a standalone event detection model. We compared its performance on the environmental sound classification ESC-50 dataset against the widely used industry models BEATS and PANN. The SenseVoice model achieved commendable results on these tasks. However, due to limitations in training data and methodology, its event classification performance has some gaps compared to specialized AED models.
91
+
92
+ <div align=""center"">
93
+ <img src=""image/aed_figure.png"" width=""500"" />
94
+ </div>
95
+
96
+ ## Computational Efficiency
97
+
98
+ The SenseVoice-Small model deploys a non-autoregressive end-to-end architecture, resulting in extremely low inference latency. With a similar number of parameters to the Whisper-Small model, it infers more than 5 times faster than Whisper-Small and 15 times faster than Whisper-Large.
99
+
100
+ <div align=""center"">
101
+ <img src=""image/inference.png"" width=""1000"" />
102
+ </div>
103
+
104
+
105
+ # Requirements
106
+
107
+ ```shell
108
+ pip install -r requirements.txt
109
+ ```
110
+
111
+ <a name=""Usage""></a>
112
+ # Usage
113
+
114
+ ## Inference
115
+
116
+ Supports input of audio in any format and of any duration.
117
+
118
+ ```python
119
+ from funasr import AutoModel
120
+ from funasr.utils.postprocess_utils import rich_transcription_postprocess
121
+
122
+ model_dir = ""FunAudioLLM/SenseVoiceSmall""
123
+
124
+
125
+ model = AutoModel(
126
+ model=model_dir,
127
+ vad_model=""fsmn-vad"",
128
+ vad_kwargs={""max_single_segment_time"": 30000},
129
+ device=""cuda:0"",
130
+ hub=""hf"",
131
+ )
132
+
133
+ # en
134
+ res = model.generate(
135
+ input=f""{model.model_path}/example/en.mp3"",
136
+ cache={},
137
+ language=""auto"", # ""zn"", ""en"", ""yue"", ""ja"", ""ko"", ""nospeech""
138
+ use_itn=True,
139
+ batch_size_s=60,
140
+ merge_vad=True, #
141
+ merge_length_s=15,
142
+ )
143
+ text = rich_transcription_postprocess(res[0][""text""])
144
+ print(text)
145
+ ```
146
+
147
+ Parameter Description:
148
+ - `model_dir`: The name of the model, or the path to the model on the local disk.
149
+ - `vad_model`: This indicates the activation of VAD (Voice Activity Detection). The purpose of VAD is to split long audio into shorter clips. In this case, the inference time includes both VAD and SenseVoice total consumption, and represents the end-to-end latency. If you wish to test the SenseVoice model's inference time separately, the VAD model can be disabled.
150
+ - `vad_kwargs`: Specifies the configurations for the VAD model. `max_single_segment_time`: denotes the maximum duration for audio segmentation by the `vad_model`, with the unit being milliseconds (ms).
151
+ - `use_itn`: Whether the output result includes punctuation and inverse text normalization.
152
+ - `batch_size_s`: Indicates the use of dynamic batching, where the total duration of audio in the batch is measured in seconds (s).
153
+ - `merge_vad`: Whether to merge short audio fragments segmented by the VAD model, with the merged length being `merge_length_s`, in seconds (s).
154
+
155
+ If all inputs are short audios (<30s), and batch inference is needed to speed up inference efficiency, the VAD model can be removed, and `batch_size` can be set accordingly.
156
+ ```python
157
+ model = AutoModel(model=model_dir, device=""cuda:0"", hub=""hf"")
158
+
159
+ res = model.generate(
160
+ input=f""{model.model_path}/example/en.mp3"",
161
+ cache={},
162
+ language=""zh"", # ""zn"", ""en"", ""yue"", ""ja"", ""ko"", ""nospeech""
163
+ use_itn=False,
164
+ batch_size=64,
165
+ hub=""hf"",
166
+ )
167
+ ```
168
+
169
+ For more usage, please refer to [docs](https://github.com/modelscope/FunASR/blob/main/docs/tutorial/README.md)
170
+
171
+ ### Inference directly
172
+
173
+ Supports input of audio in any format, with an input duration limit of 30 seconds or less.
174
+
175
+ ```python
176
+ from model import SenseVoiceSmall
177
+ from funasr.utils.postprocess_utils import rich_transcription_postprocess
178
+
179
+ model_dir = ""FunAudioLLM/SenseVoiceSmall""
180
+ m, kwargs = SenseVoiceSmall.from_pretrained(model=model_dir, device=""cuda:0"", hub=""hf"")
181
+ m.eval()
182
+
183
+ res = m.inference(
184
+ data_in=f""{kwargs['model_path']}/example/en.mp3"",
185
+ language=""auto"", # ""zn"", ""en"", ""yue"", ""ja"", ""ko"", ""nospeech""
186
+ use_itn=False,
187
+ **kwargs,
188
+ )
189
+
190
+ text = rich_transcription_postprocess(res[0][0][""text""])
191
+ print(text)
192
+ ```
193
+
194
+ ### Export and Test (*On going*)
195
+ Ref to [SenseVoice](https://github.com/FunAudioLLM/SenseVoice)
196
+ ## Service
197
+
198
+ Ref to [SenseVoice](https://github.com/FunAudioLLM/SenseVoice)
199
+
200
+ ## Finetune
201
+
202
+ Ref to [SenseVoice](https://github.com/FunAudioLLM/SenseVoice)
203
+
204
+ ## WebUI
205
+
206
+ ```shell
207
+ python webui.py
208
+ ```
209
+
210
+ <div align=""center""><img src=""image/webui.png"" width=""700""/> </div>
211
+
212
+ <a name=""Community""></a>
213
+ # Community
214
+ If you encounter problems in use, you can directly raise Issues on the github page.
215
+
216
+ You can also scan the following DingTalk group QR code to join the community group for communication and discussion.
217
+
218
+ | FunAudioLLM | FunASR |
219
+ |:----------------------------------------------------------------:|:--------------------------------------------------------:|
220
+ | <div align=""left""><img src=""image/dingding_sv.png"" width=""250""/> | <img src=""image/dingding_funasr.png"" width=""250""/></div> |","{""id"": ""FunAudioLLM/SenseVoiceSmall"", ""author"": ""FunAudioLLM"", ""sha"": ""3eb3b4eeffc2f2dde6051b853983753db33e35c3"", ""last_modified"": ""2024-07-31 05:47:48+00:00"", ""created_at"": ""2024-07-03 03:56:49+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 1092, ""downloads_all_time"": null, ""likes"": 250, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""en"", ""zh"", ""ja"", ""ko"", ""license:other"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""language:\n- en\n- zh\n- ja\n- ko\nlicense: other\nlicense_name: model-license\nlicense_link: https://github.com/modelscope/FunASR/blob/main/MODEL_LICENSE\nlibrary: funasr"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README_ja.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README_zh.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='am.mvn', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='chn_jpn_yue_eng_ko_spectok.bpe.model', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='configuration.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='demo.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='example/en.mp3', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='example/ja.mp3', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='example/ko.mp3', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='example/yue.mp3', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='example/zh.mp3', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image/aed_figure.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image/asr_results.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image/asr_results1.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image/asr_results2.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image/dingding_funasr.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image/dingding_sv.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image/inference.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image/sensevoice.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image/sensevoice2.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image/ser_figure.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image/ser_table.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image/webui.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='image/wechat.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='requirements.txt', size=None, blob_id=None, lfs=None)""], ""spaces"": [""FunAudioLLM/SenseVoice"", ""megatrump/test-FunAudioLLM"", ""terryli/cantonese-call-transcriber"", ""cuio/SenseVoice"", ""Nocigar/siliconflow"", ""megatrump/SenseVoice"", ""tuan243/checkSound""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-07-31 05:47:48+00:00"", ""cardData"": ""language:\n- en\n- zh\n- ja\n- ko\nlicense: other\nlicense_name: model-license\nlicense_link: https://github.com/modelscope/FunASR/blob/main/MODEL_LICENSE\nlibrary: funasr"", ""transformersInfo"": null, ""_id"": ""6684cc014099f9d44cd40dbd"", ""modelId"": ""FunAudioLLM/SenseVoiceSmall"", ""usedStorage"": 940388053}",0,https://huggingface.co/ChenChenyu/SenseVoiceSmall-finetuned,1,,0,,0,,0,"FunAudioLLM/SenseVoice, Nocigar/siliconflow, cuio/SenseVoice, huggingface/InferenceSupport/discussions/new?title=FunAudioLLM/SenseVoiceSmall&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BFunAudioLLM%2FSenseVoiceSmall%5D(%2FFunAudioLLM%2FSenseVoiceSmall)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, megatrump/SenseVoice, megatrump/test-FunAudioLLM, terryli/cantonese-call-transcriber, tuan243/checkSound",8
221
+ ChenChenyu/SenseVoiceSmall-finetuned,"---
222
+ datasets:
223
+ - ChenChenyu/VoiceDataSet
224
+ base_model:
225
+ - FunAudioLLM/SenseVoiceSmall
226
+ ---
227
+ # SenseVoiceSmall微调模型
228
+ 对SenseVoiceSmall开源模型用粤语和四川话数据集进行微调，得到微调后的模型。
229
+ ## 微调后测试cer
230
+ | | 带符号 | 去符号 |
231
+ | :-----| ----: | :----: |
232
+ | 微调前 | 0.35151298237809847 | 0.19419020076531499 |
233
+ | 微调后 | 0.17646826424209877 | 0.12941482715620842 |
234
+ ## 微调所用数据集
235
+ ChenChenyu/VoiceDataSet","{""id"": ""ChenChenyu/SenseVoiceSmall-finetuned"", ""author"": ""ChenChenyu"", ""sha"": ""0562b755c36b202ca07ef65b05577e7fd5a39bcf"", ""last_modified"": ""2024-12-01 11:42:52+00:00"", ""created_at"": ""2024-12-01 05:11:23+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 1, ""downloads_all_time"": null, ""likes"": 2, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""tensorboard"", ""dataset:ChenChenyu/VoiceDataSet"", ""base_model:FunAudioLLM/SenseVoiceSmall"", ""base_model:finetune:FunAudioLLM/SenseVoiceSmall"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- FunAudioLLM/SenseVoiceSmall\ndatasets:\n- ChenChenyu/VoiceDataSet"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='.ipynb_checkpoints/config-checkpoint.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='1best_recog/text', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='2024-10-25/18-23-57/.hydra/config.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='2024-10-25/18-23-57/.hydra/hydra.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='2024-10-25/18-23-57/.hydra/overrides.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='2024-10-25/18-23-57/train_ds.log', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.yaml', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='configuration.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='log.txt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.pt.best', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tensorboard/events.out.tfevents.1729851842.autodl-container-f42f45a886-bccfcaff', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-12-01 11:42:52+00:00"", ""cardData"": ""base_model:\n- FunAudioLLM/SenseVoiceSmall\ndatasets:\n- ChenChenyu/VoiceDataSet"", ""transformersInfo"": null, ""_id"": ""674beffb6563eabf0c165cd9"", ""modelId"": ""ChenChenyu/SenseVoiceSmall-finetuned"", ""usedStorage"": 5685682742}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=ChenChenyu/SenseVoiceSmall-finetuned&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BChenChenyu%2FSenseVoiceSmall-finetuned%5D(%2FChenChenyu%2FSenseVoiceSmall-finetuned)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
SmolDocling-256M-preview_finetunes_20250424_223250.csv_finetunes_20250424_223250.csv ADDED
@@ -0,0 +1,597 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ ds4sd/SmolDocling-256M-preview,"---
3
+ base_model:
4
+ - HuggingFaceTB/SmolVLM-256M-Instruct
5
+ language:
6
+ - en
7
+ library_name: transformers
8
+ license: cdla-permissive-2.0
9
+ pipeline_tag: image-text-to-text
10
+ ---
11
+
12
+ <div style=""display: flex; align-items: center;"">
13
+ <img src=""https://huggingface.co/ds4sd/SmolDocling-256M-preview/resolve/main/assets/SmolDocling_doctags1.png"" alt=""SmolDocling"" style=""width: 200px; height: auto; margin-right: 20px;"">
14
+ <div>
15
+ <h3>SmolDocling-256M-preview</h3>
16
+ <p>SmolDocling is a multimodal Image-Text-to-Text model designed for efficient document conversion. It retains Docling's most popular features while ensuring full compatibility with Docling through seamless support for <strong>DoclingDocuments</strong>.</p>
17
+ </div>
18
+ </div>
19
+
20
+ This model was presented in the paper [SmolDocling: An ultra-compact vision-language model for end-to-end multi-modal document conversion](https://huggingface.co/papers/2503.11576).
21
+
22
+ ### 🚀 Features:
23
+ - 🏷️ **DocTags for Efficient Tokenization** – Introduces DocTags an efficient and minimal representation for documents that is fully compatible with **DoclingDocuments**.
24
+ - 🔍 **OCR (Optical Character Recognition)** – Extracts text accurately from images.
25
+ - 📐 **Layout and Localization** – Preserves document structure and document element **bounding boxes**.
26
+ - 💻 **Code Recognition** – Detects and formats code blocks including identation.
27
+ - 🔢 **Formula Recognition** – Identifies and processes mathematical expressions.
28
+ - 📊 **Chart Recognition** – Extracts and interprets chart data.
29
+ - 📑 **Table Recognition** – Supports column and row headers for structured table extraction.
30
+ - 🖼️ **Figure Classification** – Differentiates figures and graphical elements.
31
+ - 📝 **Caption Correspondence** – Links captions to relevant images and figures.
32
+ - 📜 **List Grouping** – Organizes and structures list elements correctly.
33
+ - 📄 **Full-Page Conversion** – Processes entire pages for comprehensive document conversion including all page elements (code, equations, tables, charts etc.)
34
+ - 🔲 **OCR with Bounding Boxes** – OCR regions using a bounding box.
35
+ - 📂 **General Document Processing** – Trained for both scientific and non-scientific documents.
36
+ - 🔄 **Seamless Docling Integration** – Import into **Docling** and export in multiple formats.
37
+ - 💨 **Fast inference using VLLM** – Avg of 0.35 secs per page on A100 GPU.
38
+
39
+ ### 🚧 *Coming soon!*
40
+ - 📊 **Better chart recognition 🛠️**
41
+ - 📚 **One shot multi-page inference ⏱️**
42
+ - 🧪 **Chemical Recognition**
43
+ - 📙 **Datasets**
44
+
45
+ ## ⌨️ Get started (code examples)
46
+
47
+ You can use **transformers**, **vllm**, or **onnx** to perform inference, and [Docling](https://github.com/docling-project/docling) to convert results to variety of output formats (md, html, etc.):
48
+
49
+ <details>
50
+ <summary>📄 Single page image inference using Tranformers 🤖</summary>
51
+
52
+ ```python
53
+ # Prerequisites:
54
+ # pip install torch
55
+ # pip install docling_core
56
+ # pip install transformers
57
+
58
+ import torch
59
+ from docling_core.types.doc import DoclingDocument
60
+ from docling_core.types.doc.document import DocTagsDocument
61
+ from transformers import AutoProcessor, AutoModelForVision2Seq
62
+ from transformers.image_utils import load_image
63
+ from pathlib import Path
64
+
65
+ DEVICE = ""cuda"" if torch.cuda.is_available() else ""cpu""
66
+
67
+ # Load images
68
+ image = load_image(""https://upload.wikimedia.org/wikipedia/commons/7/76/GazettedeFrance.jpg"")
69
+
70
+ # Initialize processor and model
71
+ processor = AutoProcessor.from_pretrained(""ds4sd/SmolDocling-256M-preview"")
72
+ model = AutoModelForVision2Seq.from_pretrained(
73
+ ""ds4sd/SmolDocling-256M-preview"",
74
+ torch_dtype=torch.bfloat16,
75
+ _attn_implementation=""flash_attention_2"" if DEVICE == ""cuda"" else ""eager"",
76
+ ).to(DEVICE)
77
+
78
+ # Create input messages
79
+ messages = [
80
+ {
81
+ ""role"": ""user"",
82
+ ""content"": [
83
+ {""type"": ""image""},
84
+ {""type"": ""text"", ""text"": ""Convert this page to docling.""}
85
+ ]
86
+ },
87
+ ]
88
+
89
+ # Prepare inputs
90
+ prompt = processor.apply_chat_template(messages, add_generation_prompt=True)
91
+ inputs = processor(text=prompt, images=[image], return_tensors=""pt"")
92
+ inputs = inputs.to(DEVICE)
93
+
94
+ # Generate outputs
95
+ generated_ids = model.generate(**inputs, max_new_tokens=8192)
96
+ prompt_length = inputs.input_ids.shape[1]
97
+ trimmed_generated_ids = generated_ids[:, prompt_length:]
98
+ doctags = processor.batch_decode(
99
+ trimmed_generated_ids,
100
+ skip_special_tokens=False,
101
+ )[0].lstrip()
102
+
103
+ # Populate document
104
+ doctags_doc = DocTagsDocument.from_doctags_and_image_pairs([doctags], [image])
105
+ print(doctags)
106
+ # create a docling document
107
+ doc = DoclingDocument(name=""Document"")
108
+ doc.load_from_doctags(doctags_doc)
109
+
110
+ # export as any format
111
+ # HTML
112
+ # Path(""Out/"").mkdir(parents=True, exist_ok=True)
113
+ # output_path_html = Path(""Out/"") / ""example.html""
114
+ # doc.save_as_html(output_path_html)
115
+ # MD
116
+ print(doc.export_to_markdown())
117
+ ```
118
+ </details>
119
+
120
+
121
+ <details>
122
+ <summary> 🚀 Fast Batch Inference Using VLLM</summary>
123
+
124
+ ```python
125
+ # Prerequisites:
126
+ # pip install vllm
127
+ # pip install docling_core
128
+ # place page images you want to convert into ""img/"" dir
129
+
130
+ import time
131
+ import os
132
+ from vllm import LLM, SamplingParams
133
+ from PIL import Image
134
+ from docling_core.types.doc import DoclingDocument
135
+ from docling_core.types.doc.document import DocTagsDocument
136
+ from pathlib import Path
137
+
138
+ # Configuration
139
+ MODEL_PATH = ""ds4sd/SmolDocling-256M-preview""
140
+ IMAGE_DIR = ""img/"" # Place your page images here
141
+ OUTPUT_DIR = ""out/""
142
+ PROMPT_TEXT = ""Convert page to Docling.""
143
+
144
+ # Ensure output directory exists
145
+ os.makedirs(OUTPUT_DIR, exist_ok=True)
146
+
147
+ # Initialize LLM
148
+ llm = LLM(model=MODEL_PATH, limit_mm_per_prompt={""image"": 1})
149
+
150
+ sampling_params = SamplingParams(
151
+ temperature=0.0,
152
+ max_tokens=8192)
153
+
154
+ chat_template = f""<|im_start|>User:<image>{PROMPT_TEXT}<end_of_utterance>
155
+ Assistant:""
156
+
157
+ image_files = sorted([f for f in os.listdir(IMAGE_DIR) if f.lower().endswith(("".png"", "".jpg"", "".jpeg""))])
158
+
159
+ start_time = time.time()
160
+ total_tokens = 0
161
+
162
+ for idx, img_file in enumerate(image_files, 1):
163
+ img_path = os.path.join(IMAGE_DIR, img_file)
164
+ image = Image.open(img_path).convert(""RGB"")
165
+
166
+ llm_input = {""prompt"": chat_template, ""multi_modal_data"": {""image"": image}}
167
+ output = llm.generate([llm_input], sampling_params=sampling_params)[0]
168
+
169
+ doctags = output.outputs[0].text
170
+ img_fn = os.path.splitext(img_file)[0]
171
+ output_filename = img_fn + "".dt""
172
+ output_path = os.path.join(OUTPUT_DIR, output_filename)
173
+
174
+ with open(output_path, ""w"", encoding=""utf-8"") as f:
175
+ f.write(doctags)
176
+
177
+ # To convert to Docling Document, MD, HTML, etc.:
178
+ doctags_doc = DocTagsDocument.from_doctags_and_image_pairs([doctags], [image])
179
+ doc = DoclingDocument(name=""Document"")
180
+ doc.load_from_doctags(doctags_doc)
181
+ # export as any format
182
+ # HTML
183
+ # output_path_html = Path(OUTPUT_DIR) / f""{img_fn}.html""
184
+ # doc.save_as_html(output_path_html)
185
+ # MD
186
+ output_path_md = Path(OUTPUT_DIR) / f""{img_fn}.md""
187
+ doc.save_as_markdown(output_path_md)
188
+ print(f""Total time: {time.time() - start_time:.2f} sec"")
189
+ ```
190
+ </details>
191
+ <details>
192
+ <summary> ONNX Inference</summary>
193
+
194
+ ```python
195
+ # Prerequisites:
196
+ # pip install onnxruntime
197
+ # pip install onnxruntime-gpu
198
+ from transformers import AutoConfig, AutoProcessor
199
+ from transformers.image_utils import load_image
200
+ import onnxruntime
201
+ import numpy as np
202
+ import os
203
+ from docling_core.types.doc import DoclingDocument
204
+ from docling_core.types.doc.document import DocTagsDocument
205
+
206
+ os.environ[""OMP_NUM_THREADS""] = ""1""
207
+ # cuda
208
+ os.environ[""ORT_CUDA_USE_MAX_WORKSPACE""] = ""1""
209
+
210
+ # 1. Load models
211
+ ## Load config and processor
212
+ model_id = ""ds4sd/SmolDocling-256M-preview""
213
+ config = AutoConfig.from_pretrained(model_id)
214
+ processor = AutoProcessor.from_pretrained(model_id)
215
+
216
+ ## Load sessions
217
+ # !wget https://huggingface.co/ds4sd/SmolDocling-256M-preview/resolve/main/onnx/vision_encoder.onnx
218
+ # !wget https://huggingface.co/ds4sd/SmolDocling-256M-preview/resolve/main/onnx/embed_tokens.onnx
219
+ # !wget https://huggingface.co/ds4sd/SmolDocling-256M-preview/resolve/main/onnx/decoder_model_merged.onnx
220
+ # cpu
221
+ # vision_session = onnxruntime.InferenceSession(""vision_encoder.onnx"")
222
+ # embed_session = onnxruntime.InferenceSession(""embed_tokens.onnx"")
223
+ # decoder_session = onnxruntime.InferenceSession(""decoder_model_merged.onnx""
224
+
225
+ # cuda
226
+ vision_session = onnxruntime.InferenceSession(""vision_encoder.onnx"", providers=[""CUDAExecutionProvider""])
227
+ embed_session = onnxruntime.InferenceSession(""embed_tokens.onnx"", providers=[""CUDAExecutionProvider""])
228
+ decoder_session = onnxruntime.InferenceSession(""decoder_model_merged.onnx"", providers=[""CUDAExecutionProvider""])
229
+
230
+ ## Set config values
231
+ num_key_value_heads = config.text_config.num_key_value_heads
232
+ head_dim = config.text_config.head_dim
233
+ num_hidden_layers = config.text_config.num_hidden_layers
234
+ eos_token_id = config.text_config.eos_token_id
235
+ image_token_id = config.image_token_id
236
+ end_of_utterance_id = processor.tokenizer.convert_tokens_to_ids(""<end_of_utterance>"")
237
+
238
+ # 2. Prepare inputs
239
+ ## Create input messages
240
+ messages = [
241
+ {
242
+ ""role"": ""user"",
243
+ ""content"": [
244
+ {""type"": ""image""},
245
+ {""type"": ""text"", ""text"": ""Convert this page to docling.""}
246
+ ]
247
+ },
248
+ ]
249
+
250
+ ## Load image and apply processor
251
+ image = load_image(""https://ibm.biz/docling-page-with-table"")
252
+ prompt = processor.apply_chat_template(messages, add_generation_prompt=True)
253
+ inputs = processor(text=prompt, images=[image], return_tensors=""np"")
254
+
255
+ ## Prepare decoder inputs
256
+ batch_size = inputs['input_ids'].shape[0]
257
+ past_key_values = {
258
+ f'past_key_values.{layer}.{kv}': np.zeros([batch_size, num_key_value_heads, 0, head_dim], dtype=np.float32)
259
+ for layer in range(num_hidden_layers)
260
+ for kv in ('key', 'value')
261
+ }
262
+ image_features = None
263
+ input_ids = inputs['input_ids']
264
+ attention_mask = inputs['attention_mask']
265
+ position_ids = np.cumsum(inputs['attention_mask'], axis=-1)
266
+
267
+
268
+ # 3. Generation loop
269
+ max_new_tokens = 8192
270
+ generated_tokens = np.array([[]], dtype=np.int64)
271
+ for i in range(max_new_tokens):
272
+ inputs_embeds = embed_session.run(None, {'input_ids': input_ids})[0]
273
+
274
+ if image_features is None:
275
+ ## Only compute vision features if not already computed
276
+ image_features = vision_session.run(
277
+ ['image_features'], # List of output names or indices
278
+ {
279
+ 'pixel_values': inputs['pixel_values'],
280
+ 'pixel_attention_mask': inputs['pixel_attention_mask'].astype(np.bool_)
281
+ }
282
+ )[0]
283
+
284
+ ## Merge text and vision embeddings
285
+ inputs_embeds[inputs['input_ids'] == image_token_id] = image_features.reshape(-1, image_features.shape[-1])
286
+
287
+ logits, *present_key_values = decoder_session.run(None, dict(
288
+ inputs_embeds=inputs_embeds,
289
+ attention_mask=attention_mask,
290
+ position_ids=position_ids,
291
+ **past_key_values,
292
+ ))
293
+
294
+ ## Update values for next generation loop
295
+ input_ids = logits[:, -1].argmax(-1, keepdims=True)
296
+ attention_mask = np.ones_like(input_ids)
297
+ position_ids = position_ids[:, -1:] + 1
298
+ for j, key in enumerate(past_key_values):
299
+ past_key_values[key] = present_key_values[j]
300
+
301
+ generated_tokens = np.concatenate([generated_tokens, input_ids], axis=-1)
302
+ if (input_ids == eos_token_id).all() or (input_ids == end_of_utterance_id).all():
303
+ break # Stop predicting
304
+
305
+ doctags = processor.batch_decode(
306
+ generated_tokens,
307
+ skip_special_tokens=False,
308
+ )[0].lstrip()
309
+
310
+ print(doctags)
311
+
312
+ doctags_doc = DocTagsDocument.from_doctags_and_image_pairs([doctags], [image])
313
+ print(doctags)
314
+ # create a docling document
315
+ doc = DoclingDocument(name=""Document"")
316
+ doc.load_from_doctags(doctags_doc)
317
+
318
+ print(doc.export_to_markdown())
319
+ ```
320
+ </details>
321
+
322
+
323
+ 💻 Local inference on Apple Silicon with MLX: [see here](https://huggingface.co/ds4sd/SmolDocling-256M-preview-mlx-bf16)
324
+
325
+ ## DocTags
326
+
327
+ <img src=""https://huggingface.co/ds4sd/SmolDocling-256M-preview/resolve/main/assets/doctags_v2.png"" width=""800"" height=""auto"" alt=""Image description"">
328
+ DocTags create a clear and structured system of tags and rules that separate text from the document's structure. This makes things easier for Image-to-Sequence models by reducing confusion. On the other hand, converting directly to formats like HTML or Markdown can be messy—it often loses details, doesn’t clearly show the document’s layout, and increases the number of tokens, making processing less efficient.
329
+ DocTags are integrated with Docling, which allows export to HTML, Markdown, and JSON. These exports can be offloaded to the CPU, reducing token generation overhead and improving efficiency.
330
+
331
+ ## Supported Instructions
332
+
333
+ <table>
334
+ <tr>
335
+ <td><b>Description</b></td>
336
+ <td><b>Instruction</b></td>
337
+ <td><b>Comment</b></td>
338
+ </tr>
339
+ <tr>
340
+ <td><b>Full conversion</b></td>
341
+ <td>Convert this page to docling.</td>
342
+ <td>DocTags represetation</td>
343
+ </tr>
344
+ <tr>
345
+ <td><b>Chart</b></td>
346
+ <td>Convert chart to table.</td>
347
+ <td>(e.g., &lt;chart&gt;)</td>
348
+ </tr>
349
+ <tr>
350
+ <td><b>Formula</b></td>
351
+ <td>Convert formula to LaTeX.</td>
352
+ <td>(e.g., &lt;formula&gt;)</td>
353
+ </tr>
354
+ <tr>
355
+ <td><b>Code</b></td>
356
+ <td>Convert code to text.</td>
357
+ <td>(e.g., &lt;code&gt;)</td>
358
+ </tr>
359
+ <tr>
360
+ <td><b>Table</b></td>
361
+ <td>Convert table to OTSL.</td>
362
+ <td>(e.g., &lt;otsl&gt;) OTSL: <a href=""https://arxiv.org/pdf/2305.03393"">Lysak et al., 2023</a></td>
363
+ </tr>
364
+ <tr>
365
+ <td rowspan=4><b>Actions and Pipelines</b></td>
366
+ <td>OCR the text in a specific location: &lt;loc_155&gt;&lt;loc_233&gt;&lt;loc_206&gt;&lt;loc_237&gt;</td>
367
+ <td></td>
368
+ </tr>
369
+ <tr>
370
+ <td>Identify element at: &lt;loc_247&gt;&lt;loc_482&gt;&lt;10c_252&gt;&lt;loc_486&gt;</td>
371
+ <td></td>
372
+ </tr>
373
+ <tr>
374
+ <td>Find all 'text' elements on the page, retrieve all section headers.</td>
375
+ <td></td>
376
+ </tr>
377
+ <tr>
378
+ <td>Detect footer elements on the page.</td>
379
+ <td></td>
380
+ </tr>
381
+ </table>
382
+
383
+ #### Model Summary
384
+
385
+ - **Developed by:** Docling Team, IBM Research
386
+ - **Model type:** Multi-modal model (image+text)
387
+ - **Language(s) (NLP):** English
388
+ - **License:** Apache 2.0
389
+ - **Architecture:** Based on [Idefics3](https://huggingface.co/HuggingFaceM4/Idefics3-8B-Llama3) (see technical summary)
390
+ - **Finetuned from model:** Based on [SmolVLM-256M-Instruct](https://huggingface.co/HuggingFaceTB/SmolVLM-256M-Instruct)
391
+
392
+ **Repository:** [Docling](https://github.com/docling-project/docling)
393
+
394
+ **Paper:** [arXiv](https://arxiv.org/abs/2503.11576)
395
+
396
+ **Project Page:** [Hugging Face](https://huggingface.co/ds4sd/SmolDocling-256M-preview)
397
+
398
+ **Citation:**
399
+ ```
400
+ @misc{nassar2025smoldoclingultracompactvisionlanguagemodel,
401
+ title={SmolDocling: An ultra-compact vision-language model for end-to-end multi-modal document conversion},
402
+ author={Ahmed Nassar and Andres Marafioti and Matteo Omenetti and Maksym Lysak and Nikolaos Livathinos and Christoph Auer and Lucas Morin and Rafael Teixeira de Lima and Yusik Kim and A. Said Gurbuz and Michele Dolfi and Miquel Farré and Peter W. J. Staar},
403
+ year={2025},
404
+ eprint={2503.11576},
405
+ archivePrefix={arXiv},
406
+ primaryClass={cs.CV},
407
+ url={https://arxiv.org/abs/2503.11576},
408
+ }
409
+ ```
410
+ **Demo:** [HF Space](https://huggingface.co/spaces/ds4sd/SmolDocling-256M-Demo)","{""id"": ""ds4sd/SmolDocling-256M-preview"", ""author"": ""ds4sd"", ""sha"": ""492bde898f2bed6b493b4da8256c93de29e03a9b"", ""last_modified"": ""2025-03-23 10:50:10+00:00"", ""created_at"": ""2025-02-12 15:40:33+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 79570, ""downloads_all_time"": null, ""likes"": 1264, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""onnx"", ""safetensors"", ""idefics3"", ""image-text-to-text"", ""conversational"", ""en"", ""arxiv:2503.11576"", ""arxiv:2305.03393"", ""base_model:HuggingFaceTB/SmolVLM-256M-Instruct"", ""base_model:quantized:HuggingFaceTB/SmolVLM-256M-Instruct"", ""license:cdla-permissive-2.0"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""image-text-to-text"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- HuggingFaceTB/SmolVLM-256M-Instruct\nlanguage:\n- en\nlibrary_name: transformers\nlicense: cdla-permissive-2.0\npipeline_tag: image-text-to-text"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": null, ""config"": {""architectures"": [""Idefics3ForConditionalGeneration""], ""model_type"": ""idefics3"", ""processor_config"": {""chat_template"": ""<|im_start|>{% for message in messages %}{{message['role'] | capitalize}}{% if message['content'][0]['type'] == 'image' %}{{':'}}{% else %}{{': '}}{% endif %}{% for line in message['content'] %}{% if line['type'] == 'text' %}{{line['text']}}{% elif line['type'] == 'image' %}{{ '<image>' }}{% endif %}{% endfor %}<end_of_utterance>\n{% endfor %}{% if add_generation_prompt %}{{ 'Assistant:' }}{% endif %}""}, ""tokenizer_config"": {""bos_token"": ""<|im_start|>"", ""chat_template"": ""<|im_start|>{% for message in messages %}{{message['role'] | capitalize}}{% if message['content'][0]['type'] == 'image' %}{{':'}}{% else %}{{': '}}{% endif %}{% for line in message['content'] %}{% if line['type'] == 'text' %}{{line['text']}}{% elif line['type'] == 'image' %}{{ '<image>' }}{% endif %}{% endfor %}<end_of_utterance>\n{% endfor %}{% if add_generation_prompt %}{{ 'Assistant:' }}{% endif %}"", ""eos_token"": ""<|im_end|>"", ""pad_token"": ""<|im_end|>"", ""unk_token"": ""<|endoftext|>""}}, ""transformers_info"": {""auto_model"": ""AutoModelForImageTextToText"", ""custom_class"": null, ""pipeline_tag"": ""image-text-to-text"", ""processor"": ""AutoProcessor""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='added_tokens.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='assets/SmolDocling_doctags1.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='assets/doctags_v2.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='chat_template.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='merges.txt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/decoder_model_merged.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/decoder_model_merged_bnb4.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/decoder_model_merged_fp16.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/decoder_model_merged_int8.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/decoder_model_merged_q4.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/decoder_model_merged_q4f16.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/decoder_model_merged_quantized.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/decoder_model_merged_uint8.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/embed_tokens.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/embed_tokens_bnb4.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/embed_tokens_fp16.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/embed_tokens_int8.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/embed_tokens_q4.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/embed_tokens_q4f16.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/embed_tokens_quantized.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/embed_tokens_uint8.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/vision_encoder.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/vision_encoder_bnb4.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/vision_encoder_fp16.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/vision_encoder_int8.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/vision_encoder_q4.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/vision_encoder_q4f16.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/vision_encoder_quantized.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='onnx/vision_encoder_uint8.onnx', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='processor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='vocab.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='zero_to_fp32.py', size=None, blob_id=None, lfs=None)""], ""spaces"": [""ds4sd/SmolDocling-256M-Demo"", ""code27panda/SmolDocling-OCR-App"", ""chunking-ai/smoldocling-preview"", ""feras-vbrl/pdf-to-markdown-converter"", ""mansari722/smoldolcemahdi"", ""ProfessorLeVesseur/PDF_Topic_Extraction_Analysis_App""], ""safetensors"": {""parameters"": {""BF16"": 256484928}, ""total"": 256484928}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-03-23 10:50:10+00:00"", ""cardData"": ""base_model:\n- HuggingFaceTB/SmolVLM-256M-Instruct\nlanguage:\n- en\nlibrary_name: transformers\nlicense: cdla-permissive-2.0\npipeline_tag: image-text-to-text"", ""transformersInfo"": {""auto_model"": ""AutoModelForImageTextToText"", ""custom_class"": null, ""pipeline_tag"": ""image-text-to-text"", ""processor"": ""AutoProcessor""}, ""_id"": ""67acc0f1b150b8272b721c3b"", ""modelId"": ""ds4sd/SmolDocling-256M-preview"", ""usedStorage"": 6522512627}",0,"https://huggingface.co/ds4sd/SmolDocling-256M-preview-mlx-bf16, https://huggingface.co/ahishamm/SmolDocling-256M-preview-mlx-fp16, https://huggingface.co/flyingtothemoon/Model1, https://huggingface.co/generalheidari/aa, https://huggingface.co/xinxin2018/xinixin, https://huggingface.co/liuxinyu701/6",6,https://huggingface.co/Tjindustries/Amanda,1,,0,,0,"ProfessorLeVesseur/PDF_Topic_Extraction_Analysis_App, chunking-ai/smoldocling-preview, code27panda/SmolDocling-OCR-App, ds4sd/SmolDocling-256M-Demo, feras-vbrl/pdf-to-markdown-converter, huggingface/InferenceSupport/discussions/69, mansari722/smoldolcemahdi",7
411
+ ds4sd/SmolDocling-256M-preview-mlx-bf16,"---
412
+ base_model:
413
+ - ds4sd/SmolDocling-256M-preview
414
+ language:
415
+ - en
416
+ library_name: transformers
417
+ license: cdla-permissive-2.0
418
+ pipeline_tag: image-text-to-text
419
+ tags:
420
+ - mlx
421
+ ---
422
+
423
+ # SmolDocling-256M-preview-mlx-bf16
424
+ This model was converted to MLX format from [`ds4sd/SmolDocling-256M-preview`]() using mlx-vlm version **0.1.18**.
425
+ Refer to the [original model card](https://huggingface.co/ds4sd/SmolDocling-256M-preview) for more details on the model.
426
+ ## Use with mlx
427
+
428
+ ```bash
429
+ pip install -U mlx-vlm pillow docling-core
430
+ ```
431
+
432
+ ```python
433
+ # /// script
434
+ # requires-python = "">=3.12""
435
+ # dependencies = [
436
+ # ""docling-core"",
437
+ # ""mlx-vlm"",
438
+ # ""pillow"",
439
+ # ]
440
+ # ///
441
+ from io import BytesIO
442
+ from pathlib import Path
443
+ from urllib.parse import urlparse
444
+
445
+ import requests
446
+ from PIL import Image
447
+ from docling_core.types.doc import ImageRefMode
448
+ from docling_core.types.doc.document import DocTagsDocument, DoclingDocument
449
+ from mlx_vlm import load, generate
450
+ from mlx_vlm.prompt_utils import apply_chat_template
451
+ from mlx_vlm.utils import load_config, stream_generate
452
+
453
+ ## Settings
454
+ SHOW_IN_BROWSER = True # Export output as HTML and open in webbrowser.
455
+
456
+ ## Load the model
457
+ model_path = ""ds4sd/SmolDocling-256M-preview-mlx-bf16""
458
+ model, processor = load(model_path)
459
+ config = load_config(model_path)
460
+
461
+ ## Prepare input
462
+ prompt = ""Convert this page to docling.""
463
+
464
+ # image = ""https://ibm.biz/docling-page-with-list""
465
+ image = ""https://ibm.biz/docling-page-with-table""
466
+
467
+ # Load image resource
468
+ if urlparse(image).scheme != """": # it is a URL
469
+ response = requests.get(image, stream=True, timeout=10)
470
+ response.raise_for_status()
471
+ pil_image = Image.open(BytesIO(response.content))
472
+ else:
473
+ pil_image = Image.open(image)
474
+
475
+ # Apply chat template
476
+ formatted_prompt = apply_chat_template(processor, config, prompt, num_images=1)
477
+
478
+ ## Generate output
479
+ print(""DocTags: \n\n"")
480
+
481
+ output = """"
482
+ for token in stream_generate(
483
+ model, processor, formatted_prompt, [image], max_tokens=4096, verbose=False
484
+ ):
485
+ output += token.text
486
+ print(token.text, end="""")
487
+ if ""</doctag>"" in token.text:
488
+ break
489
+
490
+ print(""\n\n"")
491
+
492
+ # Populate document
493
+ doctags_doc = DocTagsDocument.from_doctags_and_image_pairs([output], [pil_image])
494
+ # create a docling document
495
+ doc = DoclingDocument(name=""SampleDocument"")
496
+ doc.load_from_doctags(doctags_doc)
497
+
498
+ ## Export as any format
499
+ # Markdown
500
+ print(""Markdown: \n\n"")
501
+ print(doc.export_to_markdown())
502
+
503
+ # HTML
504
+ if SHOW_IN_BROWSER:
505
+ import webbrowser
506
+
507
+ out_path = Path(""./output.html"")
508
+ doc.save_as_html(out_path, image_mode=ImageRefMode.EMBEDDED)
509
+ webbrowser.open(f""file:///{str(out_path.resolve())}"")
510
+
511
+ ```","{""id"": ""ds4sd/SmolDocling-256M-preview-mlx-bf16"", ""author"": ""ds4sd"", ""sha"": ""6081bbb15ee185b71dae0a59d73455643ce97ecd"", ""last_modified"": ""2025-04-22 15:17:46+00:00"", ""created_at"": ""2025-03-18 08:44:08+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 2275, ""downloads_all_time"": null, ""likes"": 44, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""safetensors"", ""smolvlm"", ""image-text-to-text"", ""mlx"", ""conversational"", ""en"", ""base_model:ds4sd/SmolDocling-256M-preview"", ""base_model:finetune:ds4sd/SmolDocling-256M-preview"", ""license:cdla-permissive-2.0"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""image-text-to-text"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- ds4sd/SmolDocling-256M-preview\nlanguage:\n- en\nlibrary_name: transformers\nlicense: cdla-permissive-2.0\npipeline_tag: image-text-to-text\ntags:\n- mlx"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": null, ""config"": {""architectures"": [""SmolVLMForConditionalGeneration""], ""model_type"": ""smolvlm"", ""processor_config"": {""chat_template"": ""<|im_start|>{% for message in messages %}{{message['role'] | capitalize}}{% if message['content'][0]['type'] == 'image' %}{{':'}}{% else %}{{': '}}{% endif %}{% for line in message['content'] %}{% if line['type'] == 'text' %}{{line['text']}}{% elif line['type'] == 'image' %}{{ '<image>' }}{% endif %}{% endfor %}<end_of_utterance>\n{% endfor %}{% if add_generation_prompt %}{{ 'Assistant:' }}{% endif %}""}, ""tokenizer_config"": {""bos_token"": ""<|im_start|>"", ""chat_template"": ""<|im_start|>{% for message in messages %}{{message['role'] | capitalize}}{% if message['content'][0]['type'] == 'image' %}{{':'}}{% else %}{{': '}}{% endif %}{% for line in message['content'] %}{% if line['type'] == 'text' %}{{line['text']}}{% elif line['type'] == 'image' %}{{ '<image>' }}{% endif %}{% endfor %}<end_of_utterance>\n{% endfor %}{% if add_generation_prompt %}{{ 'Assistant:' }}{% endif %}"", ""eos_token"": ""<|im_end|>"", ""pad_token"": ""<|im_end|>"", ""unk_token"": ""<|endoftext|>""}}, ""transformers_info"": {""auto_model"": ""AutoModelForImageTextToText"", ""custom_class"": null, ""pipeline_tag"": ""image-text-to-text"", ""processor"": ""AutoTokenizer""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='added_tokens.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='chat_template.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='merges.txt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='processor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='vocab.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='zero_to_fp32.py', size=None, blob_id=None, lfs=None)""], ""spaces"": [""EGYADMIN/SAJCO-Tabuk"", ""EGYADMIN/v3"", ""EGYADMIN/SA-SAJCOAI"", ""ProfessorLeVesseur/PDF_Topic_Extraction_Analysis_App""], ""safetensors"": {""parameters"": {""BF16"": 256484928}, ""total"": 256484928}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-04-22 15:17:46+00:00"", ""cardData"": ""base_model:\n- ds4sd/SmolDocling-256M-preview\nlanguage:\n- en\nlibrary_name: transformers\nlicense: cdla-permissive-2.0\npipeline_tag: image-text-to-text\ntags:\n- mlx"", ""transformersInfo"": {""auto_model"": ""AutoModelForImageTextToText"", ""custom_class"": null, ""pipeline_tag"": ""image-text-to-text"", ""processor"": ""AutoTokenizer""}, ""_id"": ""67d932581c09b1b4d0a44aa3"", ""modelId"": ""ds4sd/SmolDocling-256M-preview-mlx-bf16"", ""usedStorage"": 513026834}",1,,0,,0,,0,,0,"EGYADMIN/SA-SAJCOAI, EGYADMIN/SAJCO-Tabuk, EGYADMIN/v3, ProfessorLeVesseur/PDF_Topic_Extraction_Analysis_App, huggingface/InferenceSupport/discussions/new?title=ds4sd/SmolDocling-256M-preview-mlx-bf16&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bds4sd%2FSmolDocling-256M-preview-mlx-bf16%5D(%2Fds4sd%2FSmolDocling-256M-preview-mlx-bf16)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A",5
512
+ ahishamm/SmolDocling-256M-preview-mlx-fp16,"---
513
+ library_name: transformers
514
+ license: apache-2.0
515
+ language:
516
+ - en
517
+ base_model: ds4sd/SmolDocling-256M-preview
518
+ pipeline_tag: image-text-to-text
519
+ tags:
520
+ - mlx
521
+ - mlxvlm
522
+ ---
523
+
524
+ # ahishamm/SmolDocling-256M-preview-mlx-fp16
525
+ The Model [ahishamm/SmolDocling-256M-preview-mlx-fp16](https://huggingface.co/ahishamm/SmolDocling-256M-preview-mlx-fp16) was converted to MLX format from [ds4sd/SmolDocling-256M-preview](https://huggingface.co/ds4sd/SmolDocling-256M-preview)
526
+ using mlx-vlm version **0.1.17**.
527
+
528
+ ## Use with mlx
529
+ ```bash
530
+ pip install mlx-vlm
531
+ ```
532
+ ```python
533
+ from mlx_lm import load, generate
534
+ model, tokenizer = load(""ahishamm/SmolDocling-256M-preview-mlx-fp16"")
535
+ prompt = ""hello""
536
+ if hasattr(tokenizer, ""apply_chat_template"") and tokenizer.chat_template is not None:
537
+ messages = [{""role"": ""user"", ""content"": prompt}]
538
+ prompt = tokenizer.apply_chat_template(
539
+ messages, tokenize=False, add_generation_prompt=True
540
+ )
541
+ response = generate(model, tokenizer, prompt=prompt, verbose=True)
542
+ ```","{""id"": ""ahishamm/SmolDocling-256M-preview-mlx-fp16"", ""author"": ""ahishamm"", ""sha"": ""0b84106426517c1c6f3b41bd3d8f3ccc940abbed"", ""last_modified"": ""2025-03-18 08:55:32+00:00"", ""created_at"": ""2025-03-17 18:49:41+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 23, ""downloads_all_time"": null, ""likes"": 1, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""safetensors"", ""idefics3"", ""image-text-to-text"", ""mlx"", ""mlxvlm"", ""conversational"", ""en"", ""base_model:ds4sd/SmolDocling-256M-preview"", ""base_model:finetune:ds4sd/SmolDocling-256M-preview"", ""license:apache-2.0"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""image-text-to-text"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: ds4sd/SmolDocling-256M-preview\nlanguage:\n- en\nlibrary_name: transformers\nlicense: apache-2.0\npipeline_tag: image-text-to-text\ntags:\n- mlx\n- mlxvlm"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": null, ""config"": {""architectures"": [""Idefics3ForConditionalGeneration""], ""model_type"": ""idefics3"", ""processor_config"": {""chat_template"": ""<|im_start|>{% for message in messages %}{{ message['role'] | capitalize }}:{% if message['content'] is string %}{{ message['content'] }{% else %}{% if message['content'][0]['type'] == 'image' %}{{ ':' }}{% else %}{{ ': ' }}{% endif %}{% for line in message['content'] %}{% if line['type'] == 'text' %}{{ line['text'] }}{% elif line['type'] == 'image' %}{{ '<image>' }}{% endif %}{% endfor %}{% endif %}<end_of_utterance>{% endfor %}{% if add_generation_prompt %}Assistant:{% endif %}""}, ""tokenizer_config"": {""bos_token"": ""<|im_start|>"", ""chat_template"": ""<|im_start|>{% for message in messages %}{{message['role'] | capitalize}}{% if message['content'][0]['type'] == 'image' %}{{':'}}{% else %}{{': '}}{% endif %}{% for line in message['content'] %}{% if line['type'] == 'text' %}{{line['text']}}{% elif line['type'] == 'image' %}{{ '<image>' }}{% endif %}{% endfor %}<end_of_utterance>\n{% endfor %}{% if add_generation_prompt %}{{ 'Assistant:' }}{% endif %}"", ""eos_token"": ""<|im_end|>"", ""pad_token"": ""<|im_end|>"", ""unk_token"": ""<|endoftext|>""}}, ""transformers_info"": {""auto_model"": ""AutoModelForImageTextToText"", ""custom_class"": null, ""pipeline_tag"": ""image-text-to-text"", ""processor"": ""AutoProcessor""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='added_tokens.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='chat_template.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='merges.txt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='processor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='vocab.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='zero_to_fp32.py', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": {""parameters"": {""F16"": 256484928}, ""total"": 256484928}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-03-18 08:55:32+00:00"", ""cardData"": ""base_model: ds4sd/SmolDocling-256M-preview\nlanguage:\n- en\nlibrary_name: transformers\nlicense: apache-2.0\npipeline_tag: image-text-to-text\ntags:\n- mlx\n- mlxvlm"", ""transformersInfo"": {""auto_model"": ""AutoModelForImageTextToText"", ""custom_class"": null, ""pipeline_tag"": ""image-text-to-text"", ""processor"": ""AutoProcessor""}, ""_id"": ""67d86ec51b6be8ab19917e42"", ""modelId"": ""ahishamm/SmolDocling-256M-preview-mlx-fp16"", ""usedStorage"": 513026363}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=ahishamm/SmolDocling-256M-preview-mlx-fp16&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bahishamm%2FSmolDocling-256M-preview-mlx-fp16%5D(%2Fahishamm%2FSmolDocling-256M-preview-mlx-fp16)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
543
+ flyingtothemoon/Model1,"---
544
+ license: openrail
545
+ language:
546
+ - ae
547
+ base_model:
548
+ - ds4sd/SmolDocling-256M-preview
549
+ pipeline_tag: text-classification
550
+ ---","{""id"": ""flyingtothemoon/Model1"", ""author"": ""flyingtothemoon"", ""sha"": ""c7cd42a929fa577631bb39c9cfd6eddb3e1f664a"", ""last_modified"": ""2025-03-30 06:39:12+00:00"", ""created_at"": ""2025-03-30 06:38:39+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""text-classification"", ""ae"", ""base_model:ds4sd/SmolDocling-256M-preview"", ""base_model:finetune:ds4sd/SmolDocling-256M-preview"", ""license:openrail"", ""region:us""], ""pipeline_tag"": ""text-classification"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- ds4sd/SmolDocling-256M-preview\nlanguage:\n- ae\nlicense: openrail\npipeline_tag: text-classification"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-03-30 06:39:12+00:00"", ""cardData"": ""base_model:\n- ds4sd/SmolDocling-256M-preview\nlanguage:\n- ae\nlicense: openrail\npipeline_tag: text-classification"", ""transformersInfo"": null, ""_id"": ""67e8e6ef6221f5872231bd4c"", ""modelId"": ""flyingtothemoon/Model1"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=flyingtothemoon/Model1&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bflyingtothemoon%2FModel1%5D(%2Fflyingtothemoon%2FModel1)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
551
+ generalheidari/aa,"---
552
+ license: apache-2.0
553
+ language:
554
+ - aa
555
+ - en
556
+ base_model:
557
+ - ds4sd/SmolDocling-256M-preview
558
+ pipeline_tag: document-question-answering
559
+ tags:
560
+ - chemistry
561
+ datasets:
562
+ - nvidia/Llama-Nemotron-Post-Training-Dataset
563
+ metrics:
564
+ - accuracy
565
+ new_version: Qwen/Qwen2.5-Omni-7B
566
+ library_name: allennlp
567
+ ---","{""id"": ""generalheidari/aa"", ""author"": ""generalheidari"", ""sha"": ""be62e1c8dc371d46fed8398e039357d820d5ef95"", ""last_modified"": ""2025-04-14 20:46:48+00:00"", ""created_at"": ""2025-04-14 20:44:32+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""allennlp"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""allennlp"", ""chemistry"", ""document-question-answering"", ""aa"", ""en"", ""dataset:nvidia/Llama-Nemotron-Post-Training-Dataset"", ""base_model:ds4sd/SmolDocling-256M-preview"", ""base_model:finetune:ds4sd/SmolDocling-256M-preview"", ""license:apache-2.0"", ""region:us""], ""pipeline_tag"": ""document-question-answering"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- ds4sd/SmolDocling-256M-preview\ndatasets:\n- nvidia/Llama-Nemotron-Post-Training-Dataset\nlanguage:\n- aa\n- en\nlibrary_name: allennlp\nlicense: apache-2.0\nmetrics:\n- accuracy\npipeline_tag: document-question-answering\ntags:\n- chemistry\nnew_version: Qwen/Qwen2.5-Omni-7B"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-04-14 20:46:48+00:00"", ""cardData"": ""base_model:\n- ds4sd/SmolDocling-256M-preview\ndatasets:\n- nvidia/Llama-Nemotron-Post-Training-Dataset\nlanguage:\n- aa\n- en\nlibrary_name: allennlp\nlicense: apache-2.0\nmetrics:\n- accuracy\npipeline_tag: document-question-answering\ntags:\n- chemistry\nnew_version: Qwen/Qwen2.5-Omni-7B"", ""transformersInfo"": null, ""_id"": ""67fd73b0d2c9d1369d2489b2"", ""modelId"": ""generalheidari/aa"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=generalheidari/aa&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bgeneralheidari%2Faa%5D(%2Fgeneralheidari%2Faa)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
568
+ xinxin2018/xinixin,"---
569
+ license: mit
570
+ language:
571
+ - ae
572
+ metrics:
573
+ - accuracy
574
+ base_model:
575
+ - ds4sd/SmolDocling-256M-preview
576
+ new_version: black-forest-labs/FLUX.1-dev
577
+ pipeline_tag: zero-shot-classification
578
+ library_name: fasttext
579
+ tags:
580
+ - code
581
+ ---","{""id"": ""xinxin2018/xinixin"", ""author"": ""xinxin2018"", ""sha"": ""cb87aeeec9867b6d8837344731c5ffc4e094a62a"", ""last_modified"": ""2025-04-22 09:37:52+00:00"", ""created_at"": ""2025-04-22 09:37:09+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""fasttext"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""fasttext"", ""code"", ""zero-shot-classification"", ""ae"", ""base_model:ds4sd/SmolDocling-256M-preview"", ""base_model:finetune:ds4sd/SmolDocling-256M-preview"", ""license:mit"", ""region:us""], ""pipeline_tag"": ""zero-shot-classification"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- ds4sd/SmolDocling-256M-preview\nlanguage:\n- ae\nlibrary_name: fasttext\nlicense: mit\nmetrics:\n- accuracy\npipeline_tag: zero-shot-classification\ntags:\n- code\nnew_version: black-forest-labs/FLUX.1-dev"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-04-22 09:37:52+00:00"", ""cardData"": ""base_model:\n- ds4sd/SmolDocling-256M-preview\nlanguage:\n- ae\nlibrary_name: fasttext\nlicense: mit\nmetrics:\n- accuracy\npipeline_tag: zero-shot-classification\ntags:\n- code\nnew_version: black-forest-labs/FLUX.1-dev"", ""transformersInfo"": null, ""_id"": ""680763451e3d60a2b5377b0f"", ""modelId"": ""xinxin2018/xinixin"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=xinxin2018/xinixin&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bxinxin2018%2Fxinixin%5D(%2Fxinxin2018%2Fxinixin)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
582
+ liuxinyu701/6,"---
583
+ license: mit
584
+ datasets:
585
+ - nvidia/Llama-Nemotron-Post-Training-Dataset
586
+ language:
587
+ - ak
588
+ metrics:
589
+ - bertscore
590
+ base_model:
591
+ - ds4sd/SmolDocling-256M-preview
592
+ new_version: black-forest-labs/FLUX.1-dev
593
+ pipeline_tag: token-classification
594
+ library_name: fastai
595
+ tags:
596
+ - medical
597
+ ---","{""id"": ""liuxinyu701/6"", ""author"": ""liuxinyu701"", ""sha"": ""a1134495fc2efaf261a0a6f75c02db04f0c92d84"", ""last_modified"": ""2025-04-23 09:22:25+00:00"", ""created_at"": ""2025-04-23 09:22:02+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""fastai"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""fastai"", ""medical"", ""token-classification"", ""ak"", ""dataset:nvidia/Llama-Nemotron-Post-Training-Dataset"", ""base_model:ds4sd/SmolDocling-256M-preview"", ""base_model:finetune:ds4sd/SmolDocling-256M-preview"", ""license:mit"", ""region:us""], ""pipeline_tag"": ""token-classification"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- ds4sd/SmolDocling-256M-preview\ndatasets:\n- nvidia/Llama-Nemotron-Post-Training-Dataset\nlanguage:\n- ak\nlibrary_name: fastai\nlicense: mit\nmetrics:\n- bertscore\npipeline_tag: token-classification\ntags:\n- medical\nnew_version: black-forest-labs/FLUX.1-dev"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-04-23 09:22:25+00:00"", ""cardData"": ""base_model:\n- ds4sd/SmolDocling-256M-preview\ndatasets:\n- nvidia/Llama-Nemotron-Post-Training-Dataset\nlanguage:\n- ak\nlibrary_name: fastai\nlicense: mit\nmetrics:\n- bertscore\npipeline_tag: token-classification\ntags:\n- medical\nnew_version: black-forest-labs/FLUX.1-dev"", ""transformersInfo"": null, ""_id"": ""6808b13a0086d953de94fc31"", ""modelId"": ""liuxinyu701/6"", ""usedStorage"": 0}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=liuxinyu701/6&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bliuxinyu701%2F6%5D(%2Fliuxinyu701%2F6)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
SmolLM-135M_finetunes_20250426_215237.csv_finetunes_20250426_215237.csv ADDED
The diff for this file is too large to render. See raw diff
 
UAE-Large-V1_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv ADDED
The diff for this file is too large to render. See raw diff
 
Yarn-Mistral-7b-128k_finetunes_20250425_143346.csv_finetunes_20250425_143346.csv ADDED
@@ -0,0 +1,344 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ NousResearch/Yarn-Mistral-7b-128k,"---
3
+ datasets:
4
+ - emozilla/yarn-train-tokenized-16k-mistral
5
+ metrics:
6
+ - perplexity
7
+ library_name: transformers
8
+ license: apache-2.0
9
+ language:
10
+ - en
11
+ ---
12
+
13
+ # Model Card: Nous-Yarn-Mistral-7b-128k
14
+
15
+ [Preprint (arXiv)](https://arxiv.org/abs/2309.00071)
16
+ [GitHub](https://github.com/jquesnelle/yarn)
17
+ ![yarn](https://raw.githubusercontent.com/jquesnelle/yarn/mistral/data/proofpile-long-small-mistral.csv.png)
18
+
19
+ ## Model Description
20
+
21
+ Nous-Yarn-Mistral-7b-128k is a state-of-the-art language model for long context, further pretrained on long context data for 1500 steps using the YaRN extension method.
22
+ It is an extension of [Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) and supports a 128k token context window.
23
+
24
+ To use, pass `trust_remote_code=True` when loading the model, for example
25
+
26
+ ```python
27
+ model = AutoModelForCausalLM.from_pretrained(""NousResearch/Yarn-Mistral-7b-128k"",
28
+ use_flash_attention_2=True,
29
+ torch_dtype=torch.bfloat16,
30
+ device_map=""auto"",
31
+ trust_remote_code=True)
32
+ ```
33
+
34
+ In addition you will need to use the latest version of `transformers` (until 4.35 comes out)
35
+ ```sh
36
+ pip install git+https://github.com/huggingface/transformers
37
+ ```
38
+
39
+ ## Benchmarks
40
+
41
+ Long context benchmarks:
42
+ | Model | Context Window | 8k PPL | 16k PPL | 32k PPL | 64k PPL | 128k PPL |
43
+ |-------|---------------:|------:|----------:|-----:|-----:|------------:|
44
+ | [Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) | 8k | 2.96 | - | - | - | - |
45
+ | [Yarn-Mistral-7b-64k](https://huggingface.co/NousResearch/Yarn-Mistral-7b-64k) | 64k | 3.04 | 2.65 | 2.44 | 2.20 | - |
46
+ | [Yarn-Mistral-7b-128k](https://huggingface.co/NousResearch/Yarn-Mistral-7b-128k) | 128k | 3.08 | 2.68 | 2.47 | 2.24 | 2.19 |
47
+
48
+ Short context benchmarks showing that quality degradation is minimal:
49
+ | Model | Context Window | ARC-c | Hellaswag | MMLU | Truthful QA |
50
+ |-------|---------------:|------:|----------:|-----:|------------:|
51
+ | [Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) | 8k | 59.98 | 83.31 | 64.16 | 42.15 |
52
+ | [Yarn-Mistral-7b-64k](https://huggingface.co/NousResearch/Yarn-Mistral-7b-64k) | 64k | 59.38 | 81.21 | 61.32 | 42.50 |
53
+ | [Yarn-Mistral-7b-128k](https://huggingface.co/NousResearch/Yarn-Mistral-7b-128k) | 128k | 58.87 | 80.58 | 60.64 | 42.46 |
54
+
55
+ ## Collaborators
56
+
57
+ - [bloc97](https://github.com/bloc97): Methods, paper and evals
58
+ - [@theemozilla](https://twitter.com/theemozilla): Methods, paper, model training, and evals
59
+ - [@EnricoShippole](https://twitter.com/EnricoShippole): Model training
60
+ - [honglu2875](https://github.com/honglu2875): Paper and evals
61
+
62
+ The authors would like to thank LAION AI for their support of compute for this model.
63
+ It was trained on the [JUWELS](https://www.fz-juelich.de/en/ias/jsc/systems/supercomputers/juwels) supercomputer.","{""id"": ""NousResearch/Yarn-Mistral-7b-128k"", ""author"": ""NousResearch"", ""sha"": ""d09f1f8ed437d61c1aff94c1beabee554843dcdd"", ""last_modified"": ""2023-11-02 20:01:56+00:00"", ""created_at"": ""2023-10-31 13:15:14+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 10139, ""downloads_all_time"": null, ""likes"": 572, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""pytorch"", ""mistral"", ""text-generation"", ""custom_code"", ""en"", ""dataset:emozilla/yarn-train-tokenized-16k-mistral"", ""arxiv:2309.00071"", ""license:apache-2.0"", ""autotrain_compatible"", ""text-generation-inference"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""datasets:\n- emozilla/yarn-train-tokenized-16k-mistral\nlanguage:\n- en\nlibrary_name: transformers\nlicense: apache-2.0\nmetrics:\n- perplexity"", ""widget_data"": [{""text"": ""My name is Julien and I like to""}, {""text"": ""I like traveling by train because""}, {""text"": ""Paris is an amazing place to visit,""}, {""text"": ""Once upon a time,""}], ""model_index"": null, ""config"": {""architectures"": [""MistralForCausalLM""], ""auto_map"": {""AutoConfig"": ""configuration_mistral.MistralConfig"", ""AutoModelForCausalLM"": ""modeling_mistral_yarn.MistralForCausalLM""}, ""model_type"": ""mistral"", ""tokenizer_config"": {""bos_token"": ""<s>"", ""eos_token"": ""</s>"", ""pad_token"": null, ""unk_token"": ""<unk>"", ""use_default_system_prompt"": true}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='added_tokens.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='configuration_mistral.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='modeling_mistral_yarn.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00001-of-00002.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00002-of-00002.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model.bin.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.model', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [""featherless-ai/try-this-model"", ""Sarath0x8f/Document-QA-bot"", ""limcheekin/Yarn-Mistral-7B-128k-GGUF"", ""Darok/Featherless-Feud"", ""JDWebProgrammer/chatbot"", ""emekaboris/try-this-model"", ""realgenius/NousResearch-Yarn-Mistral-7b-128k"", ""ryn-85/NousResearch-Yarn-Mistral-7b-128k"", ""SC999/NV_Nemotron"", ""VKCYBER/NousResearch-Yarn-Mistral-7b-128k"", ""bradarrML/NousResearch-Yarn-Mistral-7b-128k"", ""PeepDaSlan9/NousResearch-Yarn-Mistral-7b-128k"", ""ziqin/NousResearch-Yarn-Mistral-7b-128k2"", ""kichen/NousResearch-Yarn-Mistral-7b-128k"", ""bhandsab/NousResearch-Yarn-Mistral-7b-128k"", ""Li46666/NousResearch-Yarn-Mistral-7b-128k"", ""TogetherAI/NousResearch-Yarn-Mistral-7b-128k"", ""bruc/NousResearch-Yarn-Mistral-7b-128k"", ""intelligenix/NousResearch-Yarn-Mistral-7b-128k"", ""Bellamy66/NousResearch-Yarn-Mistral-7b-128k"", ""Hboris/NousResearch-Yarn-Mistral-7b-128k"", ""harmindersinghnijjar/NousResearch-Yarn-Mistral-7b-128k"", ""schogini/NousResearch-Yarn-Mistral-7b-128k"", ""schogini/test3"", ""hijaukuohno/NousResearch-Yarn-Mistral-7b-128k"", ""JacksonGa/NousResearch-Yarn-Mistral-7b-128k"", ""ziqin/NousResearch-Yarn-Mistral-7b-128k"", ""leckneck/NousResearch-Yarn-Mistral-7b-128k"", ""Risb0v/NousResearch-Yarn-Mistral-7b-128k"", ""JackHoltone/try-this-model"", ""neelumsoft/Document-QA-bot"", ""k11112/try-this-model"", ""sailokesh/Hello_GPT""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-11-02 20:01:56+00:00"", ""cardData"": ""datasets:\n- emozilla/yarn-train-tokenized-16k-mistral\nlanguage:\n- en\nlibrary_name: transformers\nlicense: apache-2.0\nmetrics:\n- perplexity"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""_id"": ""6540fde232dcbb86631c0227"", ""modelId"": ""NousResearch/Yarn-Mistral-7b-128k"", ""usedStorage"": 28967555174}",0,"https://huggingface.co/dustydecapod/unraveled-7b-sft-lora, https://huggingface.co/dustydecapod/unraveled-7b-dpo-lora, https://huggingface.co/dustydecapod/unraveled-7b-a1, https://huggingface.co/Nitral-Archive/Kunocchini-1.2-7b-longtext-broken, https://huggingface.co/Alphatao/a91b1374-b60f-4823-9aad-5f976d07f08e",5,"https://huggingface.co/Yorth/ChatMistral128k, https://huggingface.co/xishanlone/Yarn-Mistral-7b-finetuned, https://huggingface.co/mrbmaryam/misteral-500, https://huggingface.co/mrbmaryam/misteral-100-logsummary, https://huggingface.co/codys12/Mistral-7b-Pathway-128k, https://huggingface.co/codys12/Mistral-7b-Pathway-128k-2, https://huggingface.co/codys12/Mistral-7b-Pathway-128k-4, https://huggingface.co/Shaleen123/mistrallite-medical-qa-2, https://huggingface.co/Shaleen123/nousmistral_128k_medical_qa, https://huggingface.co/Shaleen123/nousmistral_128k_medical_qa_full, https://huggingface.co/Shaleen123/nousmistral_128k_medical_qai, https://huggingface.co/Shaleen123/nousmistral_128k_medical_full, https://huggingface.co/openerotica/mistral-7b-lamia-v0.2, https://huggingface.co/eeeebbb2/4ce6d22d-2cc2-4626-9b60-d1f5d8f3141d, https://huggingface.co/dada22231/4ce6d22d-2cc2-4626-9b60-d1f5d8f3141d, https://huggingface.co/dada22231/19d6e509-260a-448f-9fe0-71b0dc32dfc2, https://huggingface.co/eeeebbb2/19d6e509-260a-448f-9fe0-71b0dc32dfc2, https://huggingface.co/eeeebbb2/1dea215e-69aa-4677-83a9-30354b142b14, https://huggingface.co/dada22231/1dea215e-69aa-4677-83a9-30354b142b14, https://huggingface.co/eeeebbb2/2c7c2cbb-c84e-4a7f-b44d-9eeba1a88832, https://huggingface.co/laquythang/2c7c2cbb-c84e-4a7f-b44d-9eeba1a88832, https://huggingface.co/dada22231/2c7c2cbb-c84e-4a7f-b44d-9eeba1a88832, https://huggingface.co/1-lock/c5aa696e-b261-4d7b-80b4-292d1a774c12, https://huggingface.co/dada22231/c5aa696e-b261-4d7b-80b4-292d1a774c12, https://huggingface.co/eeeebbb2/c5aa696e-b261-4d7b-80b4-292d1a774c12, https://huggingface.co/1-lock/cbb1f466-752c-485e-a59f-e9572b87753a, https://huggingface.co/dada22231/cbb1f466-752c-485e-a59f-e9572b87753a, https://huggingface.co/eeeebbb2/cbb1f466-752c-485e-a59f-e9572b87753a, https://huggingface.co/1-lock/0ffdac01-fb3b-4cff-a490-aee966862d58, https://huggingface.co/eeeebbb2/0ffdac01-fb3b-4cff-a490-aee966862d58, https://huggingface.co/dada22231/0ffdac01-fb3b-4cff-a490-aee966862d58, https://huggingface.co/DeepDream2045/0ffdac01-fb3b-4cff-a490-aee966862d58, https://huggingface.co/diaenra/3a0bf135-4483-4954-a86b-5ec161f4fd6f, https://huggingface.co/DeepDream2045/3a0bf135-4483-4954-a86b-5ec161f4fd6f, https://huggingface.co/vdos/3a0bf135-4483-4954-a86b-5ec161f4fd6f, https://huggingface.co/1-lock/3a0bf135-4483-4954-a86b-5ec161f4fd6f, https://huggingface.co/dada22231/3a0bf135-4483-4954-a86b-5ec161f4fd6f, https://huggingface.co/eeeebbb2/5eb46825-b54e-4c8a-bb73-60acbff95428, https://huggingface.co/DeepDream2045/5eb46825-b54e-4c8a-bb73-60acbff95428, https://huggingface.co/vdos/5eb46825-b54e-4c8a-bb73-60acbff95428, https://huggingface.co/Rodo-Sami/5eb46825-b54e-4c8a-bb73-60acbff95428, https://huggingface.co/bbytxt/5eb46825-b54e-4c8a-bb73-60acbff95428, https://huggingface.co/nblinh63/c1f32187-fe6d-48e7-8500-d8b23a1fa4a4, https://huggingface.co/nhung03/f99c37b8-5770-49ea-be1c-c88bacfc3cf5, https://huggingface.co/vdos/f99c37b8-5770-49ea-be1c-c88bacfc3cf5, https://huggingface.co/nblinh/33d3b6a7-3b08-4875-b054-69e951a0b698, https://huggingface.co/nhung03/33d3b6a7-3b08-4875-b054-69e951a0b698, https://huggingface.co/dzanbek/71075de7-094b-40a8-9370-70e796b03836, https://huggingface.co/dimasik87/7d340111-1a1d-4f5b-ba8c-3745be181c7f, https://huggingface.co/Alpha-Command-Intern/8760f2f0-9900-42e8-a6ae-77734c05cd96, https://huggingface.co/VERSIL91/f836eeb3-4606-4b4f-9640-a8bb76a06856, https://huggingface.co/VERSIL91/5d9c7aa1-6d86-4600-8206-d3b344c95963, https://huggingface.co/VERSIL91/8f458f01-1b18-4575-bb50-6412d63838e5, https://huggingface.co/VERSIL91/271de863-a7f6-4948-a696-68307f03bfbb, https://huggingface.co/fedovtt/b89a3fab-953d-446b-81e9-9adbdb1f2238, https://huggingface.co/VERSIL91/ee95b17f-8fbf-46c7-808f-51067c74af87, https://huggingface.co/VERSIL91/70aa3e8e-dc12-4654-9dc7-a8856ddd0717, https://huggingface.co/VERSIL91/e92b77ba-0846-48ad-859d-4709a193889d, https://huggingface.co/bbytxt/7f355202-7e2e-4737-9cbb-17d05cef0713, https://huggingface.co/dimasik2987/a463009c-ae0d-4a08-9a80-00926e3cfefb, https://huggingface.co/nttx/f85d5102-692f-438a-a95c-daad1d43366b, https://huggingface.co/VERSIL91/771936df-afb8-46d8-b492-6449a6e3bfb4, https://huggingface.co/cunghoctienganh/2d9bef26-5121-4a93-831b-759dc1222cfc, https://huggingface.co/nttx/cc501e30-6022-4b30-ac14-fbf2aca02b7e, https://huggingface.co/thaffggg/cc501e30-6022-4b30-ac14-fbf2aca02b7e, https://huggingface.co/VERSIL91/770e9b0b-c82a-42ab-b985-1cc3be79f64a, https://huggingface.co/tuanna08go/770e9b0b-c82a-42ab-b985-1cc3be79f64a, https://huggingface.co/VERSIL91/69f0e633-9233-41fa-82cd-62eef3c92b94, https://huggingface.co/chauhoang/69f0e633-9233-41fa-82cd-62eef3c92b94, https://huggingface.co/ivangrapher/4af1208a-ce50-4ba1-ac26-89ca6c48b25c, https://huggingface.co/lesso06/4af1208a-ce50-4ba1-ac26-89ca6c48b25c, https://huggingface.co/bbytxt/4af1208a-ce50-4ba1-ac26-89ca6c48b25c, https://huggingface.co/VERSIL91/4af1208a-ce50-4ba1-ac26-89ca6c48b25c, https://huggingface.co/tuanna08go/4bd9f0d7-ead0-4f41-a052-734fa308e920, https://huggingface.co/duyphu/82d20613-483f-75f6-7641-6f3a3dd42aea, https://huggingface.co/VERSIL91/4b28e435-fc17-4b53-9eec-37d06dcfda76, https://huggingface.co/havinash-ai/874f1b55-1b51-4b7e-a128-186d939f303a, https://huggingface.co/lesso03/233137a3-3739-4554-ab23-36f3aac99796, https://huggingface.co/lesso02/07166209-7a69-47ff-b17e-3ba8ba3c380f, https://huggingface.co/tuanna08go/bfb40f1e-a273-cc8f-3425-54157447d677, https://huggingface.co/VERSIL91/10077c3b-5443-4eea-a7ee-3d7091c6efa4, https://huggingface.co/hongngo/4b036ee2-1b8f-4ce0-b1e8-632c9d857caf, https://huggingface.co/havinash-ai/158e14a0-7af7-4b95-82e0-09a1890906b9, https://huggingface.co/tuanna08go/d997cf21-6327-8e8c-4dc8-5aa0556d28a7, https://huggingface.co/VERSIL91/f89c02e7-5cec-428a-81df-593399bb5292, https://huggingface.co/duyphu/06980322-b4ea-0d0b-14f1-d30f1b517b4b, https://huggingface.co/chauhoang/73f3fabd-9900-7f2f-a011-788b4a8a4e7a, https://huggingface.co/thangla01/bba44761-be3e-433f-ac89-abe7a58f02c8, https://huggingface.co/chauhoang/0d8825c1-4c2f-446d-b5d9-674eb35b6ab3, https://huggingface.co/nbninh/cebbdc47-52da-4a51-8767-fc8eee12c09f, https://huggingface.co/sergioalves/b97fadea-da8c-449c-9fba-a8fb645818db, https://huggingface.co/nhoxinh/a4ab1f22-ae51-47cd-8845-ba17cc673ba8, https://huggingface.co/nhung02/67274d88-1656-4507-bcb8-413b94593bfa, https://huggingface.co/havinash-ai/9364427f-10a1-42eb-a8fa-63bcc66f0d65, https://huggingface.co/prxy5606/ce496f76-5788-441a-8068-293ae481375e, https://huggingface.co/thakkkkkk/c71e0c77-e77b-43c3-92b4-8ff2929aac76, https://huggingface.co/lesso05/b69a9bc5-ff12-4cfb-a380-517001da1215, https://huggingface.co/lesso04/d573e283-70e5-43d2-b285-59c49a924313, https://huggingface.co/0x1202/54510b0f-9761-43c4-9538-84c9fd328f2e, https://huggingface.co/dzanbek/45aae6fc-ab5a-4388-ae18-f139ba4d0fe9, https://huggingface.co/demohong/ecf33ef3-1c6b-4c82-997d-8e562035dbed, https://huggingface.co/nblinh63/05013627-414d-4699-979a-3fbad8e0a362, https://huggingface.co/thalllsssss/5654d20c-136f-4866-b6f8-33b136c5d2f9, https://huggingface.co/lesso13/91e2307c-40ff-491d-8e2f-c7125442bff5, https://huggingface.co/chauhoang/f5cdacb6-c0c5-4333-835d-a56a34b48f64, https://huggingface.co/chauhoang/30ea8b1e-e86b-41be-8ead-3ad30f9185d5, https://huggingface.co/ivangrapher/193e6a11-46ee-4b3c-b06e-f0ef1875a14b, https://huggingface.co/nhung03/31f63669-8c6b-4f64-b745-17ecc3de4454, https://huggingface.co/cunghoctienganh/dcb93bcf-d036-4d7f-8491-0481b980afef, https://huggingface.co/nbninh/88576970-d870-4734-a57c-95be35433f55, https://huggingface.co/chauhoang/dcc8c1d1-bb81-dade-08a6-f71099da91bb, https://huggingface.co/sergioalves/cb5bc371-adb3-4162-99ac-bac83ab8fe30, https://huggingface.co/prxy5604/acde2aa1-da86-44e8-82ab-026ae06a22cf, https://huggingface.co/demohong/b862661f-6f25-4d75-b802-005b6aa58427, https://huggingface.co/cunghoctienganh/e11f4b17-0a59-4901-a211-d5bbcce62c24, https://huggingface.co/filipesantoscv11/f4683856-db91-457a-a0a0-66923a7d949a, https://huggingface.co/nblinh/4828ebde-6c91-46b5-ac7c-9ee9b64b373c, https://huggingface.co/hongngo/2c1c831e-fed3-4c54-846a-3d8125d3c839, https://huggingface.co/dzanbek/1faa732e-5ca7-433a-a23d-dbf0394d990f, https://huggingface.co/prxy5604/ed9ebf74-873e-490d-99f4-88ca5d9a4821, https://huggingface.co/tuanna08go/bacef684-33a2-4bef-b29c-ee533d381f26, https://huggingface.co/chauhoang/abad27d3-edac-4dfe-ab3e-c6e9f7bf5197, https://huggingface.co/kokovova/ad8ae14d-e729-43b7-9569-cf39f6b2386a, https://huggingface.co/mrHunghddddd/6f945f76-ca69-4f87-b4b6-67c6c73d6149, https://huggingface.co/thakkkkkk/a530cd42-41e2-47ed-a822-a9f184dcfe31, https://huggingface.co/datlaaaaaaa/47f518fa-0b53-44f0-985a-6b4f0da10500, https://huggingface.co/joboffer/042444c8-8bf7-4d09-8855-cc2a07443e8e, https://huggingface.co/thangla01/d24bad42-2d70-45ca-acab-1176699ab624, https://huggingface.co/kk-aivio/da92bc1f-5e73-4173-97f2-5c21135e2632, https://huggingface.co/duyphu/a6c251b1-a0d4-4526-a47c-b6b54b9bde22, https://huggingface.co/kokovova/cf99fa8b-d16f-4c15-a58a-bb723a8cda39, https://huggingface.co/nttx/298bb42d-4012-404d-9f8f-eb6048249fc1, https://huggingface.co/ajtaltarabukin2022/76ab40f5-5c46-4b75-a0bc-6b102d3fd437, https://huggingface.co/mrHunghddddd/d5009a75-3a93-4137-bc48-56e5bcf8a828, https://huggingface.co/trangtrannnnn/52892534-3c5f-40fd-89de-61e4c746cbaa, https://huggingface.co/mrHungddddh/1d8f7a65-db6b-4220-b38e-0a0fac876d13, https://huggingface.co/nhung03/9ff30e73-0e2b-46da-b2b4-318b02ff0406, https://huggingface.co/nbninh/f943b7fe-faab-4d30-8358-95ba05f5705c, https://huggingface.co/sergioalves/d5c49a01-bc31-40d5-93dd-c43c3227769e, https://huggingface.co/aleegis10/3ce4e187-e2fd-4d0b-a953-39d335ed509f, https://huggingface.co/Aivesa/3ac905c6-b56d-4d35-810a-855512cee15e, https://huggingface.co/thalllsssss/992804ea-dc84-41df-95fb-a4a652d54e24, https://huggingface.co/cunghoctienganh/22f11baf-fab2-497a-9b32-c23e68bb6fdf, https://huggingface.co/nhunglaaaaaaa/6a00ab3f-6749-41f9-97a2-689f35710cfd, https://huggingface.co/nhung03/76525ae0-d4de-48eb-b830-4aa2c9b98537, https://huggingface.co/vertings6/0a301c11-8510-4a89-aaa5-9997e278d52a, https://huggingface.co/nhung01/09b12016-4de8-415e-b542-23da6221d45e, https://huggingface.co/lesso01/fde5de22-97a5-4bfe-9a01-aec6b3c73899, https://huggingface.co/ClarenceDan/ee762543-2342-4a42-bc24-b76bf89d72f6, https://huggingface.co/nttx/9652d64d-a000-4fc3-8d60-f8c656a3850e, https://huggingface.co/0x1202/38773a2e-6cf0-49f4-8cf4-6e34bf6e0a44, https://huggingface.co/0x1202/0b652084-976b-4171-9147-545c4823b511, https://huggingface.co/aleegis10/29e624ab-bd2f-4fd1-8cda-b0607541dd27, https://huggingface.co/prxy5608/b5f89c95-3ff8-4e68-bbfc-92eb3f9d916c, https://huggingface.co/kokovova/18f82232-bedc-4aea-8ee3-0f3a38a70fc3, https://huggingface.co/mrHunghddddd/7f550d89-48b9-4b61-a234-3aea758c63c4, https://huggingface.co/trangtrannnnn/0df8fdfd-bb45-4ec3-8dce-ca012b1fcb38, https://huggingface.co/vermoney/edfc823e-4960-402a-9d15-b2334cc9c541, https://huggingface.co/dimasik2987/4b4a07fe-8d4b-42bc-9da2-94b82ef498f9, https://huggingface.co/lesso04/2a35ee11-e5b5-4b0b-b08d-168779525e1c, https://huggingface.co/thakkkkkk/0ac3a980-9432-4c64-9805-82d82405de05, https://huggingface.co/myhaaaaaaa/89c7dba5-b650-46d4-8cc5-62fc8d80cd18, https://huggingface.co/nhung02/3fb51f48-e6e8-47c6-b330-29ec61639cd1, https://huggingface.co/tuanna08go/fd51f0b4-3619-4421-86a8-479621440cb0, https://huggingface.co/lesso17/51831be6-c9e4-4d01-bf19-ffe768726467, https://huggingface.co/nhunglaaaaaaa/aedd1742-0ec5-4d24-bc8b-3cb237ec53bb, https://huggingface.co/prxy5605/4234cbbc-7d58-41fc-9dcd-11e61cfe0c16, https://huggingface.co/mrHunghddddd/0bfed24f-0502-4868-910d-8d1c3d81c4f7, https://huggingface.co/hongngo/cf7ca3f6-c50e-426c-a6d2-c8851b7f7904, https://huggingface.co/laquythang/80211eee-f0bb-4b5d-abf2-edad88a75ab3, https://huggingface.co/tuanna08go/11aa8c68-e283-4ef6-a298-be9fd84ecdf7, https://huggingface.co/nhunglaaaaaaa/6989f3aa-2b1e-49b1-acc7-40795eb271e7, https://huggingface.co/lesso09/344aae17-8f2a-4d84-8922-406e07dd82bf, https://huggingface.co/lesso06/f9c54df4-fcfe-4f01-b216-59537b091b8c, https://huggingface.co/daniel40/0a08a1f4-9596-47ab-be0d-a8a5ba12cde8, https://huggingface.co/mrhunghd/41c28e2c-0cff-4303-957b-f763c2764195, https://huggingface.co/thalllsssss/11624806-f966-42dd-ac76-209541255c05, https://huggingface.co/prxy5606/b1ba74b4-b098-4fba-8013-414a2ec3deb2, https://huggingface.co/duyphu/b373ad09-9cab-4a56-b398-870664f5384d, https://huggingface.co/lesso16/8a6496a8-9041-433a-8697-e9ab66b8c9f3, https://huggingface.co/kokovova/ec62609a-b24a-43ce-90f5-8bb74c81b826, https://huggingface.co/kk-aivio/dd104cd7-7b24-4f93-884f-3e516660a1ec, https://huggingface.co/chauhoang/dd722a00-e3df-423b-b84f-9d9d41fd03b5, https://huggingface.co/vmpsergio/444ea8e2-879b-4a96-a165-93a22dadcfe1, https://huggingface.co/aleegis09/a1f0bf08-c29a-41bc-9220-affbd0c4cadd, https://huggingface.co/robiual-awal/38659625-f1ff-44a5-ac8a-90bc2c08a907, https://huggingface.co/lesso15/bd4ba09f-3628-4741-93b3-06a475e3758d, https://huggingface.co/JacksonBrune/db47e4c0-cec7-46c8-9e99-114552080598, https://huggingface.co/prxy5608/2d62c95b-b5b7-4cff-b313-4fd6817fc10c, https://huggingface.co/sergioalves/591bc345-d926-4863-ba0d-fa3094334212, https://huggingface.co/infogeo/c52aef5b-e8e5-4af4-b6e7-9b623ff9403c, https://huggingface.co/robiual-awal/8fecbd4a-3cb4-4e10-96da-54e457c7a8f9, https://huggingface.co/daniel40/e9c7a482-c721-4a32-b27c-dfe64225e9b5, https://huggingface.co/prxy5605/981de68c-8e87-473e-962f-56110deecee5, https://huggingface.co/fedovtt/daebeeb1-e9d0-45fa-9a40-ef360ed3b699, https://huggingface.co/ClarenceDan/8e6e00e7-e5f2-4df3-856c-be0c4995ceef, https://huggingface.co/ClarenceDan/ef40f1bf-ad6e-4b5c-a188-7f8c76d6991e, https://huggingface.co/tuanna08go/cdefd8e3-48f1-4dc3-8bb7-3d8126ab5c3e, https://huggingface.co/duyphu/3207f062-dabf-4c6d-a787-260a00d67134, https://huggingface.co/cvoffer/46e8d4c9-682e-4a6f-b008-d8c016d6061c, https://huggingface.co/dimasik1987/c0f2fe3c-8d6d-4391-bb5c-13d8dbd23854, https://huggingface.co/lesso16/3403cb0a-afbb-4fed-93b7-cc950245ec9c, https://huggingface.co/JacksonBrune/4ee7ac79-f226-4d84-b38a-fc4495cdfb15, https://huggingface.co/dimasik1987/17b0a914-a867-4fdf-9f23-b30b63d5c5be, https://huggingface.co/kk-aivio/a31317fd-eda1-45ac-a021-4a5740c0e1e5, https://huggingface.co/prxy5604/f993da58-3631-4086-a986-0c08510aae7b, https://huggingface.co/chauhoang/0536300e-fd95-4a8a-ab92-7fa76ab0ae5b, https://huggingface.co/prxy5605/5a560af8-c547-417a-99ea-eccfb62a7d4b, https://huggingface.co/prxy5607/bcc24a39-0c77-4a8a-be1e-236284cf42e5, https://huggingface.co/aleegis10/c92f3cbc-704a-43bf-a1cf-4c08533fa0cd, https://huggingface.co/prxy5608/ea79ccf6-48bf-4045-9040-448baaebc5bd, https://huggingface.co/prxy5604/5e431d85-fe97-4246-a462-763743f55e83, https://huggingface.co/ivangrapher/11031ac8-8d03-437a-9853-b63de2765100, https://huggingface.co/lesso11/883bf879-f6bc-4098-83a4-ca1597cb56df, https://huggingface.co/great0001/5c60bc32-dfd6-47b4-a126-90849066cc87, https://huggingface.co/great0001/28bc450b-afdc-47c8-b615-650b703d91d2, https://huggingface.co/duyphu/02428a4b-60ec-4f62-ba81-1c46f99962ea, https://huggingface.co/lesso08/9769c5df-7858-437e-bb65-b4b3686ad21a, https://huggingface.co/robiulawaldev/fc68744a-dc32-404d-8547-8ad21a5eacd5, https://huggingface.co/daniel40/e2f1684c-464c-4dee-b698-08ba6243be57, https://huggingface.co/havinash-ai/de5599dd-765d-4873-8bc1-ab7b67b4d954, https://huggingface.co/lesso09/58bf2eef-085b-4529-be7d-4ab58745fecc, https://huggingface.co/lesso01/25241cfe-2462-42c6-803f-93306b9ae111, https://huggingface.co/shibajustfor/a690348b-9d15-46c5-92f0-37a6633c8cd5, https://huggingface.co/lesso05/3d6c2bef-5196-4e10-a18c-e8a671e5592b, https://huggingface.co/havinash-ai/04d3c6eb-b8cd-4219-be2d-b4d5b8f18e47, https://huggingface.co/prxy5604/718a3229-73ba-4980-9425-ffb241853776, https://huggingface.co/lesso17/508f660c-12f4-4089-9dde-fe35f712951b, https://huggingface.co/shibajustfor/af3c529b-0bda-4a87-a3aa-f1f8497ef553, https://huggingface.co/lesso12/88801676-1167-4bba-9701-ae4ea2f02008, https://huggingface.co/robiulawaldev/c2b9832c-7f1f-49f7-8593-e0b85a1f902f, https://huggingface.co/adammandic87/cd1c7f2c-1a8d-40d0-9d77-30d633a1c5a2, https://huggingface.co/baby-dev/26612dc7-1f71-4054-8593-4d0debb48528, https://huggingface.co/daniel40/75886df6-0dee-44cf-a2a1-b5197d0b55ef, https://huggingface.co/baby-dev/ea97ea81-4e0b-45ba-9258-976a240c75a7, https://huggingface.co/baby-dev/ee9e4270-6e49-4849-b82c-e093f3f8f91e, https://huggingface.co/prxy5604/3f62ddfe-6a37-4444-b41d-01f4dd20cb8b, https://huggingface.co/baby-dev/06260df1-8d5f-4438-abfa-b35ff90eb2f3, https://huggingface.co/cilooor/2cb730ae-3a65-489a-b4eb-5f6a3e363c56, https://huggingface.co/lesso01/a72da487-f41d-4f09-a703-5bab3ad6c5ea, https://huggingface.co/robiulawaldev/f51fc536-ec44-4ee6-86aa-63f55f95a32d, https://huggingface.co/Best000/d6ef5ef4-583d-4099-94b0-9e06ea8ebd83, https://huggingface.co/lesso10/38cac83b-96f1-4d90-b4d5-c34c58ba5cfd, https://huggingface.co/havinash-ai/9b797226-7145-4136-93e8-a54e4d40f964, https://huggingface.co/robiulawaldev/5917997a-3b70-4b3d-9960-6615b0d995f9, https://huggingface.co/lesso17/6b390319-bab5-44be-943f-aa0dc3786961, https://huggingface.co/aleegis12/35493a02-5126-4b7f-91b7-d8871a341d0e, https://huggingface.co/lesso02/fe6121c5-10ea-4fed-b951-b5de3475ed1e, https://huggingface.co/aleegis12/58c6f737-4f6f-4799-93b0-b0649e76c322, https://huggingface.co/robiulawaldev/9970cf47-6dce-400b-a000-44ff1cccb491, https://huggingface.co/adammandic87/21cb04a5-3d13-4602-9789-69b16ae1e600, https://huggingface.co/robiual-awal/49d8bce3-7595-4775-abe3-3e5bad9cca94, https://huggingface.co/Best000/c35e532b-b7af-4878-bc4d-5e040418cbd5, https://huggingface.co/daniel40/96011ee3-48dc-4faf-bdd5-e3d9dfbf34f6, https://huggingface.co/adammandic87/b6839351-6117-4fc4-a71b-36e4e7f939cf, https://huggingface.co/robiulawaldev/515a1e0b-f80b-4231-b5ce-9676ac23b041, https://huggingface.co/shibajustfor/9f972555-a6f1-4612-a66e-32411fddee31, https://huggingface.co/havinash-ai/24827954-6ddd-4643-be09-7a4274d99af4, https://huggingface.co/robiulawaldev/b9dc2b35-928b-46aa-b46a-314a2b196a96, https://huggingface.co/adammandic87/eb6979e6-fa66-43a1-a422-43d348a52bab, https://huggingface.co/havinash-ai/957e24e1-c430-480c-9fc7-fd05fc81db31, https://huggingface.co/lesso/70f5ea81-56a7-4e34-9650-4bf1396a028e, https://huggingface.co/daniel40/f58047a9-8efb-45ce-a110-788eee8ae7cb, https://huggingface.co/lesso/11d552f9-f055-4a51-acfd-23e825d55757, https://huggingface.co/lesso/b55601b3-6885-44f3-aef3-85f4a63cdfae, https://huggingface.co/lesso/8e1c0543-8bf8-4de3-aedb-cd41f2e1f01e, https://huggingface.co/daniel40/808e0612-0f02-4f38-9adf-4549212f9fb4, https://huggingface.co/lesso/ba4b6b81-b218-46c9-a712-bf5912442432, https://huggingface.co/lesso/13d81c38-3ccf-4cd0-a680-62f3c8bf3d71, https://huggingface.co/cilooor/4fea0006-2355-4d84-8016-6fe7357801f3, https://huggingface.co/Best000/951b3595-639b-44af-b0b4-dfb8eef72989, https://huggingface.co/robiulawaldev/88374eba-2d62-47f8-9f63-2657719284eb, https://huggingface.co/shibajustfor/f64a629b-925b-4d9a-b89d-c755f66f7127, https://huggingface.co/robiulawaldev/c9932886-e440-4e7b-8f17-afe72cc16b10, https://huggingface.co/daniel40/e92e73c2-4192-479d-8dd6-08df12f89aa5, https://huggingface.co/lesso/3b9dd9df-231d-4f0c-8f91-5dcc06b68653, https://huggingface.co/abaddon182/d98b7989-907f-4509-beca-ea59427efec9, https://huggingface.co/cimol/37d9ad71-75ff-42f3-95ae-10ece10c325c, https://huggingface.co/mrferr3t/91571574-8b09-48f0-91a1-30ef8b351462, https://huggingface.co/mrferr3t/0638f241-005f-4e9c-83d7-3190ffb2520f, https://huggingface.co/lesso09/68ed5539-cf7d-48b5-99ca-180cd43d546b, https://huggingface.co/daniel40/0e9bdb7f-9568-4961-95bf-82f3481df390, https://huggingface.co/daniel40/adf8f84c-6b4d-4b4c-9321-42ba4bdea166, https://huggingface.co/lesso03/57495c52-6fe4-4495-96c8-d66a7c6b2f20, https://huggingface.co/lesso04/13f27876-66d6-4a00-a75e-419b37426b93, https://huggingface.co/tuantmdev/685940b0-6137-4dea-873e-f54936302b99, https://huggingface.co/Romain-XV/8ce2d2b2-df8e-49c2-991a-52d0e53001e8, https://huggingface.co/lesso10/5061a162-8467-4be0-85b7-d25a59052c89, https://huggingface.co/lesso02/0ee4622d-03da-45da-9673-593d4b80d202, https://huggingface.co/daniel40/d71796aa-aad8-430b-a505-3e744e207c58, https://huggingface.co/lesso16/42634c65-8919-4860-a2f5-59ee089a7350, https://huggingface.co/lesso12/9e468510-d87b-40a1-98d2-bd8a4ae92c3f, https://huggingface.co/lesso07/879c4381-fa71-48d1-b161-9dc77df0e5f8, https://huggingface.co/lesso17/4d6e571b-31ed-40e3-8bcb-39534d93d9eb, https://huggingface.co/lesso06/635f8758-9726-4773-a4aa-fc6c65d5c541, https://huggingface.co/lesso04/6b06da8d-889f-4c51-837d-35f633f20b24, https://huggingface.co/tuantmdev/23168cb6-d6ed-450f-ad61-e2d41152acb0, https://huggingface.co/trenden/0a5f8c35-a43b-45cf-9635-e3eb16b06155, https://huggingface.co/shibajustfor/5a67588d-8c19-4508-a315-f804ea005cef, https://huggingface.co/romainnn/0291b31e-c29d-4447-9085-6c8c73860149, https://huggingface.co/samoline/ea23503c-3030-4bf2-a6e0-3dc60e740ec7, https://huggingface.co/lesso13/25a03dfb-e1a7-4cc7-b1be-b2abfde71185, https://huggingface.co/Romain-XV/ff1365cc-4661-4224-a4cd-7d7f5781fc86, https://huggingface.co/lesso12/6035f9dc-356b-4106-8ff6-29e39af82047, https://huggingface.co/lesso03/343ddc57-13fe-47f2-b2e5-e07e22b97afa, https://huggingface.co/Alphatao/45822166-6d97-4b65-bb48-c9b6067f220d, https://huggingface.co/lesso11/39b4169b-a411-43ff-864d-3a134ba5fcf1, https://huggingface.co/lesso13/8ed84472-3d75-4840-a2de-c7b8c763aafb, https://huggingface.co/lesso04/07236733-cf62-42cb-9f6a-f5b38e681f6f, https://huggingface.co/lesso07/8e2ab3db-3c60-4c9d-935c-3f8e0ab7c154, https://huggingface.co/fedovtt/748e72ab-d705-4782-9859-b4740c95eeed, https://huggingface.co/dzanbek/cb751569-1861-407f-945e-06a115915807, https://huggingface.co/vmpsergio/3aac1da8-4740-4e80-af8a-c438a4677455",313,"https://huggingface.co/TheBloke/Yarn-Mistral-7B-128k-AWQ, https://huggingface.co/TheBloke/Yarn-Mistral-7B-128k-GGUF, https://huggingface.co/TheBloke/Yarn-Mistral-7B-128k-GPTQ, https://huggingface.co/Lewdiculous/Kunocchini-1.2-7b-longtext-GGUF-Imatrix, https://huggingface.co/mradermacher/Yarn-Mistral-7b-128k-GGUF, https://huggingface.co/mradermacher/Yarn-Mistral-7b-128k-i1-GGUF, https://huggingface.co/PrunaAI/NousResearch-Yarn-Mistral-7b-128k-GGUF-smashed",7,"https://huggingface.co/Aryanne/YarnLake-Swap-7B, https://huggingface.co/InnerI/InnerILLM-OpenPipe-Nous-Yarn-Mistral-optimized-1228-7B-slerp, https://huggingface.co/InnerI/InnerILLM-0x00d0-7B-slerp, https://huggingface.co/bartowski/Yarncules-7b-128k-exl2, https://huggingface.co/seyf1elislam/WestKunai-Hermes-long-128k-test-7b, https://huggingface.co/mvpmaster/openchat-3.5-0106-128k-DPO-fixed-32000, https://huggingface.co/Kukedlc/NeuralContext-7b-v1, https://huggingface.co/Kukedlc/NeuralContext-7b-v2, https://huggingface.co/ehristoforu/0001, https://huggingface.co/ehristoforu/Gistral-16B, https://huggingface.co/ehristoforu/Gistral-16B-Q4_K_M-GGUF, https://huggingface.co/ehristoforu/0000mxs, https://huggingface.co/ehristoforu/flm-m01, https://huggingface.co/ehristoforu/pm-v0.1, https://huggingface.co/ehristoforu/pm-v0.2",15,"Darok/Featherless-Feud, JDWebProgrammer/chatbot, PeepDaSlan9/NousResearch-Yarn-Mistral-7b-128k, SC999/NV_Nemotron, Sarath0x8f/Document-QA-bot, VKCYBER/NousResearch-Yarn-Mistral-7b-128k, emekaboris/try-this-model, featherless-ai/try-this-model, huggingface/InferenceSupport/discussions/new?title=NousResearch/Yarn-Mistral-7b-128k&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BNousResearch%2FYarn-Mistral-7b-128k%5D(%2FNousResearch%2FYarn-Mistral-7b-128k)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, limcheekin/Yarn-Mistral-7B-128k-GGUF, neelumsoft/Document-QA-bot, realgenius/NousResearch-Yarn-Mistral-7b-128k, ryn-85/NousResearch-Yarn-Mistral-7b-128k",13
64
+ dustydecapod/unraveled-7b-sft-lora,"---
65
+ license: apache-2.0
66
+ base_model: NousResearch/Yarn-Mistral-7b-128k
67
+ tags:
68
+ - generated_from_trainer
69
+ model-index:
70
+ - name: unraveled-7b-sft-lora
71
+ results: []
72
+ ---
73
+
74
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
75
+ should probably proofread and complete it, then remove this comment. -->
76
+
77
+ # unraveled-7b-sft-lora
78
+
79
+ This model is a fine-tuned version of [NousResearch/Yarn-Mistral-7b-128k](https://huggingface.co/NousResearch/Yarn-Mistral-7b-128k) on an unknown dataset.
80
+ It achieves the following results on the evaluation set:
81
+ - Loss: 1.0261
82
+
83
+ ## Model description
84
+
85
+ More information needed
86
+
87
+ ## Intended uses & limitations
88
+
89
+ More information needed
90
+
91
+ ## Training and evaluation data
92
+
93
+ More information needed
94
+
95
+ ## Training procedure
96
+
97
+ ### Training hyperparameters
98
+
99
+ The following hyperparameters were used during training:
100
+ - learning_rate: 2e-05
101
+ - train_batch_size: 4
102
+ - eval_batch_size: 8
103
+ - seed: 42
104
+ - distributed_type: multi-GPU
105
+ - num_devices: 4
106
+ - gradient_accumulation_steps: 128
107
+ - total_train_batch_size: 2048
108
+ - total_eval_batch_size: 32
109
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
110
+ - lr_scheduler_type: cosine
111
+ - num_epochs: 1
112
+
113
+ ### Training results
114
+
115
+ | Training Loss | Epoch | Step | Validation Loss |
116
+ |:-------------:|:-----:|:----:|:---------------:|
117
+ | 1.0242 | 0.67 | 68 | 1.0262 |
118
+
119
+
120
+ ### Framework versions
121
+
122
+ - Transformers 4.35.0
123
+ - Pytorch 2.1.0+cu118
124
+ - Datasets 2.14.6
125
+ - Tokenizers 0.14.1
126
+ ","{""id"": ""dustydecapod/unraveled-7b-sft-lora"", ""author"": ""dustydecapod"", ""sha"": ""425fbe42c81e962df3cbacb794e5fcb5f3253f89"", ""last_modified"": ""2023-11-22 09:14:14+00:00"", ""created_at"": ""2023-11-21 09:53:31+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 1, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""tensorboard"", ""safetensors"", ""mistral"", ""text-generation"", ""generated_from_trainer"", ""conversational"", ""custom_code"", ""base_model:NousResearch/Yarn-Mistral-7b-128k"", ""base_model:finetune:NousResearch/Yarn-Mistral-7b-128k"", ""license:apache-2.0"", ""autotrain_compatible"", ""text-generation-inference"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: NousResearch/Yarn-Mistral-7b-128k\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: unraveled-7b-sft-lora\n results: []"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": [{""name"": ""unraveled-7b-sft-lora"", ""results"": []}], ""config"": {""architectures"": [""MistralForCausalLM""], ""auto_map"": {""AutoConfig"": ""NousResearch/Yarn-Mistral-7b-128k--configuration_mistral.MistralConfig"", ""AutoModelForCausalLM"": ""NousResearch/Yarn-Mistral-7b-128k--modeling_mistral_yarn.MistralForCausalLM""}, ""model_type"": ""mistral"", ""tokenizer_config"": {""bos_token"": ""<s>"", ""chat_template"": ""{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n' + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}"", ""eos_token"": ""</s>"", ""pad_token"": ""</s>"", ""unk_token"": ""<unk>"", ""use_default_system_prompt"": true}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='all_results.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='eval_results.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='runs/Nov22_06-20-04_401fa5a8015d/events.out.tfevents.1700634336.401fa5a8015d.8816.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='runs/Nov22_06-30-36_401fa5a8015d/events.out.tfevents.1700634659.401fa5a8015d.9073.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='runs/Nov22_06-30-36_401fa5a8015d/events.out.tfevents.1700644438.401fa5a8015d.9073.1', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='train_results.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='trainer_state.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='training_args.bin', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-11-22 09:14:14+00:00"", ""cardData"": ""base_model: NousResearch/Yarn-Mistral-7b-128k\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: unraveled-7b-sft-lora\n results: []"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""_id"": ""655c7e1b2d1922b228a382ea"", ""modelId"": ""dustydecapod/unraveled-7b-sft-lora"", ""usedStorage"": 218155852}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=dustydecapod/unraveled-7b-sft-lora&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bdustydecapod%2Funraveled-7b-sft-lora%5D(%2Fdustydecapod%2Funraveled-7b-sft-lora)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
127
+ dustydecapod/unraveled-7b-dpo-lora,"---
128
+ license: apache-2.0
129
+ base_model: NousResearch/Yarn-Mistral-7b-128k
130
+ tags:
131
+ - generated_from_trainer
132
+ model-index:
133
+ - name: unraveled-7b-dpo-lora
134
+ results: []
135
+ ---
136
+
137
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
138
+ should probably proofread and complete it, then remove this comment. -->
139
+
140
+ # unraveled-7b-dpo-lora
141
+
142
+ This model is a fine-tuned version of [NousResearch/Yarn-Mistral-7b-128k](https://huggingface.co/NousResearch/Yarn-Mistral-7b-128k) on the None dataset.
143
+ It achieves the following results on the evaluation set:
144
+ - Loss: 0.5895
145
+ - Rewards/chosen: 0.1439
146
+ - Rewards/rejected: -0.1833
147
+ - Rewards/accuracies: 0.6880
148
+ - Rewards/margins: 0.3272
149
+ - Logps/rejected: -221.8329
150
+ - Logps/chosen: -266.1414
151
+ - Logits/rejected: -1.9675
152
+ - Logits/chosen: -2.0859
153
+
154
+ ## Model description
155
+
156
+ More information needed
157
+
158
+ ## Intended uses & limitations
159
+
160
+ More information needed
161
+
162
+ ## Training and evaluation data
163
+
164
+ More information needed
165
+
166
+ ## Training procedure
167
+
168
+ ### Training hyperparameters
169
+
170
+ The following hyperparameters were used during training:
171
+ - learning_rate: 5e-07
172
+ - train_batch_size: 2
173
+ - eval_batch_size: 4
174
+ - seed: 42
175
+ - distributed_type: multi-GPU
176
+ - num_devices: 4
177
+ - gradient_accumulation_steps: 32
178
+ - total_train_batch_size: 256
179
+ - total_eval_batch_size: 16
180
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
181
+ - lr_scheduler_type: linear
182
+ - lr_scheduler_warmup_ratio: 0.1
183
+ - num_epochs: 3
184
+
185
+ ### Training results
186
+
187
+ | Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
188
+ |:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
189
+ | 0.6313 | 1.0 | 242 | 0.6318 | 0.1228 | -0.0304 | 0.6600 | 0.1532 | -220.3036 | -266.3521 | -1.9863 | -2.1062 |
190
+ | 0.6013 | 2.0 | 484 | 0.5983 | 0.1484 | -0.1334 | 0.6760 | 0.2819 | -221.3338 | -266.0959 | -1.9723 | -2.0914 |
191
+ | 0.5889 | 3.0 | 726 | 0.5895 | 0.1439 | -0.1833 | 0.6880 | 0.3272 | -221.8329 | -266.1414 | -1.9675 | -2.0859 |
192
+
193
+
194
+ ### Framework versions
195
+
196
+ - Transformers 4.35.0
197
+ - Pytorch 2.1.0+cu118
198
+ - Datasets 2.14.6
199
+ - Tokenizers 0.14.1
200
+ ","{""id"": ""dustydecapod/unraveled-7b-dpo-lora"", ""author"": ""dustydecapod"", ""sha"": ""0463ac7751cac7a16d4d22de6faa04c48ded3fb8"", ""last_modified"": ""2023-11-22 15:14:30+00:00"", ""created_at"": ""2023-11-22 09:31:42+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 1, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""tensorboard"", ""safetensors"", ""mistral"", ""text-generation"", ""generated_from_trainer"", ""conversational"", ""custom_code"", ""base_model:NousResearch/Yarn-Mistral-7b-128k"", ""base_model:finetune:NousResearch/Yarn-Mistral-7b-128k"", ""license:apache-2.0"", ""autotrain_compatible"", ""text-generation-inference"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: NousResearch/Yarn-Mistral-7b-128k\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: unraveled-7b-dpo-lora\n results: []"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": [{""name"": ""unraveled-7b-dpo-lora"", ""results"": []}], ""config"": {""architectures"": [""MistralForCausalLM""], ""auto_map"": {""AutoConfig"": ""NousResearch/Yarn-Mistral-7b-128k--configuration_mistral.MistralConfig"", ""AutoModelForCausalLM"": ""NousResearch/Yarn-Mistral-7b-128k--modeling_mistral_yarn.MistralForCausalLM""}, ""model_type"": ""mistral"", ""tokenizer_config"": {""bos_token"": ""<s>"", ""chat_template"": ""{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n' + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}"", ""eos_token"": ""</s>"", ""pad_token"": ""</s>"", ""unk_token"": ""<unk>"", ""use_default_system_prompt"": true}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='all_results.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='eval_results.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='runs/Nov22_09-29-27_401fa5a8015d/events.out.tfevents.1700645508.401fa5a8015d.9788.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='runs/Nov22_09-29-27_401fa5a8015d/events.out.tfevents.1700666050.401fa5a8015d.9788.1', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='train_results.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='trainer_state.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='training_args.bin', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-11-22 15:14:30+00:00"", ""cardData"": ""base_model: NousResearch/Yarn-Mistral-7b-128k\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: unraveled-7b-dpo-lora\n results: []"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""_id"": ""655dca7e68c6dd1321499779"", ""modelId"": ""dustydecapod/unraveled-7b-dpo-lora"", ""usedStorage"": 218197580}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=dustydecapod/unraveled-7b-dpo-lora&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bdustydecapod%2Funraveled-7b-dpo-lora%5D(%2Fdustydecapod%2Funraveled-7b-dpo-lora)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
201
+ dustydecapod/unraveled-7b-a1,"---
202
+ license: apache-2.0
203
+ base_model: NousResearch/Yarn-Mistral-7b-128k
204
+ tags:
205
+ - generated_from_trainer
206
+ model-index:
207
+ - name: unraveled-7b-dpo-lora
208
+ results: []
209
+ ---
210
+
211
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
212
+ should probably proofread and complete it, then remove this comment. -->
213
+
214
+ # unraveled-7b-dpo-lora
215
+
216
+ This model is a fine-tuned version of [NousResearch/Yarn-Mistral-7b-128k](https://huggingface.co/NousResearch/Yarn-Mistral-7b-128k), following the Zephyr alignment protocol.
217
+ It achieves the following results on the evaluation set:
218
+ - Loss: 0.5895
219
+ - Rewards/chosen: 0.1439
220
+ - Rewards/rejected: -0.1833
221
+ - Rewards/accuracies: 0.6880
222
+ - Rewards/margins: 0.3272
223
+ - Logps/rejected: -221.8329
224
+ - Logps/chosen: -266.1414
225
+ - Logits/rejected: -1.9675
226
+ - Logits/chosen: -2.0859
227
+
228
+ ## Model description
229
+
230
+ More information needed
231
+
232
+ ## Intended uses & limitations
233
+
234
+ More information needed
235
+
236
+ ## Training and evaluation data
237
+
238
+ More information needed
239
+
240
+ ## Training procedure
241
+
242
+ ### Training hyperparameters
243
+
244
+ The following hyperparameters were used during training:
245
+ - learning_rate: 5e-07
246
+ - train_batch_size: 2
247
+ - eval_batch_size: 4
248
+ - seed: 42
249
+ - distributed_type: multi-GPU
250
+ - num_devices: 4
251
+ - gradient_accumulation_steps: 32
252
+ - total_train_batch_size: 256
253
+ - total_eval_batch_size: 16
254
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
255
+ - lr_scheduler_type: linear
256
+ - lr_scheduler_warmup_ratio: 0.1
257
+ - num_epochs: 3
258
+
259
+ ### Training results
260
+
261
+ | Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
262
+ |:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
263
+ | 0.6313 | 1.0 | 242 | 0.6318 | 0.1228 | -0.0304 | 0.6600 | 0.1532 | -220.3036 | -266.3521 | -1.9863 | -2.1062 |
264
+ | 0.6013 | 2.0 | 484 | 0.5983 | 0.1484 | -0.1334 | 0.6760 | 0.2819 | -221.3338 | -266.0959 | -1.9723 | -2.0914 |
265
+ | 0.5889 | 3.0 | 726 | 0.5895 | 0.1439 | -0.1833 | 0.6880 | 0.3272 | -221.8329 | -266.1414 | -1.9675 | -2.0859 |
266
+
267
+
268
+ ### Framework versions
269
+
270
+ - Transformers 4.35.0
271
+ - Pytorch 2.1.0+cu118
272
+ - Datasets 2.14.6
273
+ - Tokenizers 0.14.1
274
+ ","{""id"": ""dustydecapod/unraveled-7b-a1"", ""author"": ""dustydecapod"", ""sha"": ""fac05775fa8121b58cda8031b7001323bd43983d"", ""last_modified"": ""2023-11-22 17:41:31+00:00"", ""created_at"": ""2023-11-22 15:25:19+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 72, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""safetensors"", ""mistral"", ""text-generation"", ""generated_from_trainer"", ""custom_code"", ""base_model:NousResearch/Yarn-Mistral-7b-128k"", ""base_model:finetune:NousResearch/Yarn-Mistral-7b-128k"", ""license:apache-2.0"", ""autotrain_compatible"", ""text-generation-inference"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: NousResearch/Yarn-Mistral-7b-128k\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: unraveled-7b-dpo-lora\n results: []"", ""widget_data"": [{""text"": ""My name is Julien and I like to""}, {""text"": ""I like traveling by train because""}, {""text"": ""Paris is an amazing place to visit,""}, {""text"": ""Once upon a time,""}], ""model_index"": [{""name"": ""unraveled-7b-dpo-lora"", ""results"": []}], ""config"": {""architectures"": [""MistralForCausalLM""], ""auto_map"": {""AutoConfig"": ""NousResearch/Yarn-Mistral-7b-128k--configuration_mistral.MistralConfig"", ""AutoModelForCausalLM"": ""NousResearch/Yarn-Mistral-7b-128k--modeling_mistral_yarn.MistralForCausalLM""}, ""model_type"": ""mistral"", ""tokenizer_config"": {""bos_token"": ""<s>"", ""eos_token"": ""</s>"", ""pad_token"": null, ""unk_token"": ""<unk>"", ""use_default_system_prompt"": true}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00001-of-00006.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00002-of-00006.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00003-of-00006.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00004-of-00006.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00005-of-00006.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00006-of-00006.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.model', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": {""parameters"": {""F32"": 7241732096}, ""total"": 7241732096}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-11-22 17:41:31+00:00"", ""cardData"": ""base_model: NousResearch/Yarn-Mistral-7b-128k\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: unraveled-7b-dpo-lora\n results: []"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""_id"": ""655e1d5f759563867174499c"", ""modelId"": ""dustydecapod/unraveled-7b-a1"", ""usedStorage"": 28966962016}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=dustydecapod/unraveled-7b-a1&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bdustydecapod%2Funraveled-7b-a1%5D(%2Fdustydecapod%2Funraveled-7b-a1)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
275
+ https://huggingface.co/Nitral-Archive/Kunocchini-1.2-7b-longtext-broken,N/A,N/A,1,,0,,0,,0,,0,,0
276
+ Alphatao/a91b1374-b60f-4823-9aad-5f976d07f08e,"---
277
+ base_model: NousResearch/Yarn-Mistral-7b-128k
278
+ library_name: transformers
279
+ model_name: a91b1374-b60f-4823-9aad-5f976d07f08e
280
+ tags:
281
+ - generated_from_trainer
282
+ - axolotl
283
+ - dpo
284
+ - trl
285
+ licence: license
286
+ ---
287
+
288
+ # Model Card for a91b1374-b60f-4823-9aad-5f976d07f08e
289
+
290
+ This model is a fine-tuned version of [NousResearch/Yarn-Mistral-7b-128k](https://huggingface.co/NousResearch/Yarn-Mistral-7b-128k).
291
+ It has been trained using [TRL](https://github.com/huggingface/trl).
292
+
293
+ ## Quick start
294
+
295
+ ```python
296
+ from transformers import pipeline
297
+
298
+ question = ""If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?""
299
+ generator = pipeline(""text-generation"", model=""Alphatao/a91b1374-b60f-4823-9aad-5f976d07f08e"", device=""cuda"")
300
+ output = generator([{""role"": ""user"", ""content"": question}], max_new_tokens=128, return_full_text=False)[0]
301
+ print(output[""generated_text""])
302
+ ```
303
+
304
+ ## Training procedure
305
+
306
+ [<img src=""https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg"" alt=""Visualize in Weights & Biases"" width=""150"" height=""24""/>](https://wandb.ai/alphatao-alphatao/Gradients-On-Demand/runs/al93galt)
307
+
308
+ This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
309
+
310
+ ### Framework versions
311
+
312
+ - TRL: 0.12.0.dev0
313
+ - Transformers: 4.46.0
314
+ - Pytorch: 2.5.0+cu124
315
+ - Datasets: 3.0.1
316
+ - Tokenizers: 0.20.1
317
+
318
+ ## Citations
319
+
320
+ Cite DPO as:
321
+
322
+ ```bibtex
323
+ @inproceedings{rafailov2023direct,
324
+ title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}},
325
+ author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn},
326
+ year = 2023,
327
+ booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023},
328
+ url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html},
329
+ editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine},
330
+ }
331
+ ```
332
+
333
+ Cite TRL as:
334
+
335
+ ```bibtex
336
+ @misc{vonwerra2022trl,
337
+ title = {{TRL: Transformer Reinforcement Learning}},
338
+ author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
339
+ year = 2020,
340
+ journal = {GitHub repository},
341
+ publisher = {GitHub},
342
+ howpublished = {\url{https://github.com/huggingface/trl}}
343
+ }
344
+ ```","{""id"": ""Alphatao/a91b1374-b60f-4823-9aad-5f976d07f08e"", ""author"": ""Alphatao"", ""sha"": ""a156ade398049a616e2de649d17d91da3245d17b"", ""last_modified"": ""2025-04-24 11:08:11+00:00"", ""created_at"": ""2025-04-24 07:41:33+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 2, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""pytorch"", ""tensorboard"", ""safetensors"", ""mistral"", ""text-generation"", ""generated_from_trainer"", ""axolotl"", ""dpo"", ""trl"", ""conversational"", ""custom_code"", ""arxiv:2305.18290"", ""base_model:NousResearch/Yarn-Mistral-7b-128k"", ""base_model:finetune:NousResearch/Yarn-Mistral-7b-128k"", ""autotrain_compatible"", ""text-generation-inference"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: NousResearch/Yarn-Mistral-7b-128k\nlibrary_name: transformers\nmodel_name: a91b1374-b60f-4823-9aad-5f976d07f08e\ntags:\n- generated_from_trainer\n- axolotl\n- dpo\n- trl\nlicence: license"", ""widget_data"": [{""text"": ""Hi, what can you help me with?""}, {""text"": ""What is 84 * 3 / 2?""}, {""text"": ""Tell me an interesting fact about the universe!""}, {""text"": ""Explain quantum computing in simple terms.""}], ""model_index"": null, ""config"": {""architectures"": [""MistralForCausalLM""], ""auto_map"": {""AutoConfig"": ""NousResearch/Yarn-Mistral-7b-128k--configuration_mistral.MistralConfig"", ""AutoModelForCausalLM"": ""NousResearch/Yarn-Mistral-7b-128k--modeling_mistral_yarn.MistralForCausalLM""}, ""model_type"": ""mistral"", ""tokenizer_config"": {""bos_token"": ""<s>"", ""chat_template"": ""{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}"", ""eos_token"": ""</s>"", ""pad_token"": ""</s>"", ""unk_token"": ""<unk>"", ""use_default_system_prompt"": true}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00001-of-00004.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00002-of-00004.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00003-of-00004.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00004-of-00004.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model.bin.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='runs/Apr24_07-41-17_f44850f1953c/events.out.tfevents.1745480507.f44850f1953c.260.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.model', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='training_args.bin', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-04-24 11:08:11+00:00"", ""cardData"": ""base_model: NousResearch/Yarn-Mistral-7b-128k\nlibrary_name: transformers\nmodel_name: a91b1374-b60f-4823-9aad-5f976d07f08e\ntags:\n- generated_from_trainer\n- axolotl\n- dpo\n- trl\nlicence: license"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""_id"": ""6809eb2d01de4afad6323264"", ""modelId"": ""Alphatao/a91b1374-b60f-4823-9aad-5f976d07f08e"", ""usedStorage"": 20532159292}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=Alphatao/a91b1374-b60f-4823-9aad-5f976d07f08e&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BAlphatao%2Fa91b1374-b60f-4823-9aad-5f976d07f08e%5D(%2FAlphatao%2Fa91b1374-b60f-4823-9aad-5f976d07f08e)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
YuE-s1-7B-anneal-en-cot_finetunes_20250426_014322.csv_finetunes_20250426_014322.csv ADDED
@@ -0,0 +1,133 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ m-a-p/YuE-s1-7B-anneal-en-cot,"---
3
+ license: apache-2.0
4
+ language:
5
+ - en
6
+ pipeline_tag: text-generation
7
+ tags:
8
+ - music
9
+ - art
10
+ - text-generation-inference
11
+ ---
12
+
13
+
14
+ <p align=""center"">
15
+ <img src=""./assets/logo/白底.png"" width=""400"" />
16
+ </p>
17
+
18
+ <p align=""center"">
19
+ <a href=""https://map-yue.github.io/"">Demo 🎶</a> &nbsp;|&nbsp; 📑 <a href=""https://arxiv.org/abs/2503.08638"">Paper</a>
20
+ <br>
21
+ <a href=""https://huggingface.co/m-a-p/YuE-s1-7B-anneal-en-cot"">YuE-s1-7B-anneal-en-cot 🤗</a> &nbsp;|&nbsp; <a href=""https://huggingface.co/m-a-p/YuE-s1-7B-anneal-en-icl"">YuE-s1-7B-anneal-en-icl 🤗</a> &nbsp;|&nbsp; <a href=""https://huggingface.co/m-a-p/YuE-s1-7B-anneal-jp-kr-cot"">YuE-s1-7B-anneal-jp-kr-cot 🤗</a>
22
+ <br>
23
+ <a href=""https://huggingface.co/m-a-p/YuE-s1-7B-anneal-jp-kr-icl"">YuE-s1-7B-anneal-jp-kr-icl 🤗</a> &nbsp;|&nbsp; <a href=""https://huggingface.co/m-a-p/YuE-s1-7B-anneal-zh-cot"">YuE-s1-7B-anneal-zh-cot 🤗</a> &nbsp;|&nbsp; <a href=""https://huggingface.co/m-a-p/YuE-s1-7B-anneal-zh-icl"">YuE-s1-7B-anneal-zh-icl 🤗</a>
24
+ <br>
25
+ <a href=""https://huggingface.co/m-a-p/YuE-s2-1B-general"">YuE-s2-1B-general 🤗</a> &nbsp;|&nbsp; <a href=""https://huggingface.co/m-a-p/YuE-upsampler"">YuE-upsampler 🤗</a>
26
+ </p>
27
+
28
+
29
+ ---
30
+ Our model's name is **YuE (乐)**. In Chinese, the word means ""music"" and ""happiness."" Some of you may find words that start with Yu hard to pronounce. If so, you can just call it ""yeah."" We wrote a song with our model's name.
31
+
32
+ <audio controls src=""https://cdn-uploads.huggingface.co/production/uploads/6555e8d8a0c34cd61a6b9ce3/rG-ELxMyzDU7zH-inB9DV.mpga""></audio>
33
+
34
+ YuE is a groundbreaking series of open-source foundation models designed for music generation, specifically for transforming lyrics into full songs (lyrics2song). It can generate a complete song, lasting several minutes, that includes both a catchy vocal track and accompaniment track. YuE is capable of modeling diverse genres/languages/vocal techniques. Please visit the [**Demo Page**](https://map-yue.github.io/) for amazing vocal performance.
35
+
36
+ Pop:Quiet Evening
37
+ <audio controls src=""https://cdn-uploads.huggingface.co/production/uploads/5fd6f670053c8345eddc1b68/HbZfODWrBK9DQyOzT3vL4.mpga""></audio>
38
+ Metal: Step Back
39
+ <audio controls src=""https://cdn-uploads.huggingface.co/production/uploads/6555e8d8a0c34cd61a6b9ce3/kmCwl4GRS70UYDEELL-Tn.mpga""></audio>
40
+ <!-- **YuE** is a groundbreaking series of open-source foundation models designed for music generation, specifically for transforming lyrics into full songs (**lyrics2song**). It can generate a complete song, lasting several minutes, that includes both a catchy vocal track and complementary accompaniment, ensuring a polished and cohesive result. -->
41
+
42
+ ## News and Updates
43
+ * 📌 Join Us on Discord! [<img alt=""join discord"" src=""https://img.shields.io/discord/842440537755353128?color=%237289da&logo=discord""/>](https://discord.gg/ssAyWMnMzu)
44
+
45
+ * **2025.03.12 🔥 Paper Released🎉**: We now release [YuE technical report](https://arxiv.org/abs/2503.08638)!!! We discuss all the technical details, findings, and lessons learned. Enjoy, and feel free to cite us~
46
+ * **2025.03.11 🫶** Now YuE supports incremental song generation!!! See [YuE-UI by joeljuvel](https://github.com/joeljuvel/YuE-UI). YuE-UI is a Gradio-based interface supporting batch generation, output selection, and continuation. You can flexibly experiment with audio prompts and different model settings, visualize your progress on an interactive timeline, rewind actions, quickly preview audio outputs at stage 1 before committing to refinement, and fully save/load your sessions (JSON format). Optimized to run smoothly even on GPUs with just 8GB VRAM using quantized models.
47
+ * **2025.02.17 🫶** Now YuE supports music continuation and Google Colab! See [YuE-extend by Mozer](https://github.com/Mozer/YuE-extend).
48
+ * **2025.02.07 🎉** Get YuE for Windows on [pinokio](https://pinokio.computer).
49
+
50
+ * **2025.01.30 🔥 Inference Update**: We now support dual-track ICL mode! You can prompt the model with a reference song, and it will generate a new song in a similar style (voice cloning [demo by @abrakjamson](https://x.com/abrakjamson/status/1885932885406093538), music style transfer [demo by @cocktailpeanut](https://x.com/cocktailpeanut/status/1886456240156348674), etc.). Try it out! 🔥🔥🔥 P.S. Be sure to check out the demos first—they're truly impressive.
51
+
52
+ * **2025.01.30 🔥 Announcement: A New Era Under Apache 2.0 🔥**: We are thrilled to announce that, in response to overwhelming requests from our community, **YuE** is now officially licensed under the **Apache 2.0** license. We sincerely hope this marks a watershed moment—akin to what Stable Diffusion and LLaMA have achieved in their respective fields—for music generation and creative AI. 🎉🎉🎉
53
+
54
+ * **2025.01.29 🎉**: We have updated the license description. we **ENCOURAGE** artists and content creators to sample and incorporate outputs generated by our model into their own works, and even monetize them. The only requirement is to credit our name: **YuE by HKUST/M-A-P** (alphabetic order).
55
+ * **2025.01.28 🫶**: Thanks to Fahd for creating a tutorial on how to quickly get started with YuE. Here is his [demonstration](https://www.youtube.com/watch?v=RSMNH9GitbA).
56
+ * **2025.01.26 🔥**: We have released the **YuE** series.
57
+
58
+
59
+ ## Quickstart
60
+
61
+ Please check our [github](https://github.com/multimodal-art-projection/YuE.git) for easy quickstart.
62
+
63
+ ---
64
+
65
+ ## License Agreement \& Disclaimer
66
+ - The YuE model (including its weights) is now released under the **Apache License, Version 2.0**. We do not make any profit from this model, and we hope it can be used for the betterment of human creativity.
67
+ - **Use & Attribution**:
68
+ - We encourage artists and content creators to freely incorporate outputs generated by YuE into their own works, including commercial projects.
69
+ - We encourage attribution to the model’s name (“YuE by HKUST/M-A-P”), especially for public and commercial use.
70
+ - **Originality & Plagiarism**: It is the sole responsibility of creators to ensure that their works, derived from or inspired by YuE outputs, do not plagiarize or unlawfully reproduce existing material. We strongly urge users to perform their own due diligence to avoid copyright infringement or other legal violations.
71
+ - **Recommended Labeling**: When uploading works to streaming platforms or sharing them publicly, we **recommend** labeling them with terms such as: “AI-generated”, “YuE-generated"", “AI-assisted” or “AI-auxiliated”. This helps maintain transparency about the creative process.
72
+ - **Disclaimer of Liability**:
73
+ - We do not assume any responsibility for the misuse of this model, including (but not limited to) illegal, malicious, or unethical activities.
74
+ - Users are solely responsible for any content generated using the YuE model and for any consequences arising from its use.
75
+ - By using this model, you agree that you understand and comply with all applicable laws and regulations regarding your generated content.
76
+
77
+ ---
78
+
79
+ ## Acknowledgements
80
+ The project is co-lead by HKUST and M-A-P (alphabetic order). Also thanks moonshot.ai, bytedance, 01.ai, and geely for supporting the project.
81
+ A friendly link to HKUST Audio group's [huggingface space](https://huggingface.co/HKUSTAudio).
82
+
83
+ We deeply appreciate all the support we received along the way. Long live open-source AI!
84
+
85
+ <br>
86
+
87
+ ## Citation
88
+
89
+ If you find our paper and code useful in your research, please consider giving a star :star: and citation :pencil: :)
90
+
91
+ ```BibTeX
92
+ @misc{yuan2025yuescalingopenfoundation,
93
+ title={YuE: Scaling Open Foundation Models for Long-Form Music Generation},
94
+ author={Ruibin Yuan and Hanfeng Lin and Shuyue Guo and Ge Zhang and Jiahao Pan and Yongyi Zang and Haohe Liu and Yiming Liang and Wenye Ma and Xingjian Du and Xinrun Du and Zhen Ye and Tianyu Zheng and Yinghao Ma and Minghao Liu and Zeyue Tian and Ziya Zhou and Liumeng Xue and Xingwei Qu and Yizhi Li and Shangda Wu and Tianhao Shen and Ziyang Ma and Jun Zhan and Chunhui Wang and Yatian Wang and Xiaowei Chi and Xinyue Zhang and Zhenzhu Yang and Xiangzhou Wang and Shansong Liu and Lingrui Mei and Peng Li and Junjie Wang and Jianwei Yu and Guojian Pang and Xu Li and Zihao Wang and Xiaohuan Zhou and Lijun Yu and Emmanouil Benetos and Yong Chen and Chenghua Lin and Xie Chen and Gus Xia and Zhaoxiang Zhang and Chao Zhang and Wenhu Chen and Xinyu Zhou and Xipeng Qiu and Roger Dannenberg and Jiaheng Liu and Jian Yang and Wenhao Huang and Wei Xue and Xu Tan and Yike Guo},
95
+ year={2025},
96
+ eprint={2503.08638},
97
+ archivePrefix={arXiv},
98
+ primaryClass={eess.AS},
99
+ url={https://arxiv.org/abs/2503.08638},
100
+ }
101
+
102
+ @misc{yuan2025yue,
103
+ title={YuE: Open Music Foundation Models for Full-Song Generation},
104
+ author={Ruibin Yuan and Hanfeng Lin and Shawn Guo and Ge Zhang and Jiahao Pan and Yongyi Zang and Haohe Liu and Xingjian Du and Xeron Du and Zhen Ye and Tianyu Zheng and Yinghao Ma and Minghao Liu and Lijun Yu and Zeyue Tian and Ziya Zhou and Liumeng Xue and Xingwei Qu and Yizhi Li and Tianhao Shen and Ziyang Ma and Shangda Wu and Jun Zhan and Chunhui Wang and Yatian Wang and Xiaohuan Zhou and Xiaowei Chi and Xinyue Zhang and Zhenzhu Yang and Yiming Liang and Xiangzhou Wang and Shansong Liu and Lingrui Mei and Peng Li and Yong Chen and Chenghua Lin and Xie Chen and Gus Xia and Zhaoxiang Zhang and Chao Zhang and Wenhu Chen and Xinyu Zhou and Xipeng Qiu and Roger Dannenberg and Jiaheng Liu and Jian Yang and Stephen Huang and Wei Xue and Xu Tan and Yike Guo},
105
+ howpublished={\url{https://github.com/multimodal-art-projection/YuE}},
106
+ year={2025},
107
+ note={GitHub repository}
108
+ }
109
+ ```
110
+ <br>","{""id"": ""m-a-p/YuE-s1-7B-anneal-en-cot"", ""author"": ""m-a-p"", ""sha"": ""454c20e1748888800f8e4b3da45125f55482d967"", ""last_modified"": ""2025-03-12 07:07:19+00:00"", ""created_at"": ""2025-01-26 16:10:21+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 21637, ""downloads_all_time"": null, ""likes"": 400, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""safetensors"", ""llama"", ""music"", ""art"", ""text-generation-inference"", ""text-generation"", ""en"", ""arxiv:2503.08638"", ""license:apache-2.0"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""language:\n- en\nlicense: apache-2.0\npipeline_tag: text-generation\ntags:\n- music\n- art\n- text-generation-inference"", ""widget_data"": [{""text"": ""My name is Julien and I like to""}, {""text"": ""I like traveling by train because""}, {""text"": ""Paris is an amazing place to visit,""}, {""text"": ""Once upon a time,""}], ""model_index"": null, ""config"": {""architectures"": [""LlamaForCausalLM""], ""model_type"": ""llama""}, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='assets/logo/\u767d\u5e95.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='assets/logo/\u900f\u660e\u5e95\u9ed1\u7ebf.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='assets/logo/\u9ed1\u5e95.png', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00001-of-00003.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00002-of-00003.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00003-of-00003.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.model', size=None, blob_id=None, lfs=None)""], ""spaces"": [""fffiloni/YuE"", ""ginipick/OpenSUNO"", ""innova-ai/YuE-music-generator-demo"", ""KBaba7/Quant"", ""bhaskartripathi/LLM_Quantization"", ""totolook/Quant"", ""FallnAI/Quantize-HF-Models"", ""ruslanmv/convert_to_gguf"", ""Davuks/OpenSUNO"", ""seawolf2357/YuE-music-generator-demo-zero"", ""VivekVKBKumar/Testing"", ""VivekVKBKumar/Testing2"", ""svjack/YuE"", ""inQuestAI/YuE"", ""multimodalart/yue-jobs"", ""AviCo/YuE"", ""Nymbo/YuE"", ""Srijal99/YuE"", ""Leryich/OpenSUNO"", ""dnola/YuE"", ""Dejansimic/YuE"", ""Harveyu/YuE-music-generator-demo"", ""ler06/YuE"", ""K00B404/LLM_Quantization"", ""toh32/YuE-music-generator-demo""], ""safetensors"": {""parameters"": {""BF16"": 6224613376}, ""total"": 6224613376}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-03-12 07:07:19+00:00"", ""cardData"": ""language:\n- en\nlicense: apache-2.0\npipeline_tag: text-generation\ntags:\n- music\n- art\n- text-generation-inference"", ""transformersInfo"": null, ""_id"": ""67965e6d9f57883759a839dd"", ""modelId"": ""m-a-p/YuE-s1-7B-anneal-en-cot"", ""usedStorage"": 12451022490}",0,https://huggingface.co/Doctor-Shotgun/YuE-s1-7B-anneal-en-cot-exl2,1,,0,"https://huggingface.co/Aryanne/YuE-s1-7B-anneal-en-cot-Q6_K-GGUF, https://huggingface.co/multimodalart/YuE-s1-7B-anneal-en-cot-Q8_0-GGUF, https://huggingface.co/DevQuasar/m-a-p.YuE-s1-7B-anneal-en-cot-GGUF, https://huggingface.co/Alissonerdx/YuE-s1-7B-anneal-en-cot-int8, https://huggingface.co/NikolayKozloff/YuE-s1-7B-anneal-en-cot-Q8_0-GGUF, https://huggingface.co/Alissonerdx/YuE-s1-7B-anneal-en-cot-nf4, https://huggingface.co/mradermacher/YuE-s1-7B-anneal-en-cot-GGUF, https://huggingface.co/mradermacher/YuE-s1-7B-anneal-en-cot-i1-GGUF, https://huggingface.co/tensorblock/YuE-s1-7B-anneal-en-cot-GGUF, https://huggingface.co/Zuellni/YuE-s1-7B-anneal-en-cot-8.0bpw-h8-exl2, https://huggingface.co/msyukorai/YuE-s1-7B-anneal-en-cot-Q4_0-GGUF, https://huggingface.co/siouni/YuE-s1-7B-anneal-en-cot-onnx-q4, https://huggingface.co/mingz2022/YuE-s1-7B-anneal-en-cot-Q4-mlx, https://huggingface.co/mingz2022/YuE-s1-7B-anneal-en-cot-Q8-mlx",14,,0,"Davuks/OpenSUNO, FallnAI/Quantize-HF-Models, K00B404/LLM_Quantization, KBaba7/Quant, Leryich/OpenSUNO, bhaskartripathi/LLM_Quantization, fffiloni/YuE, ginipick/OpenSUNO, huggingface/InferenceSupport/discussions/349, innova-ai/YuE-music-generator-demo, ruslanmv/convert_to_gguf, seawolf2357/YuE-music-generator-demo-zero, totolook/Quant",13
111
+ Doctor-Shotgun/YuE-s1-7B-anneal-en-cot-exl2,"---
112
+ quantized_by: Doctor-Shotgun
113
+ license: apache-2.0
114
+ language:
115
+ - en
116
+ tags:
117
+ - music
118
+ - llama
119
+ base_model: m-a-p/YuE-s1-7B-anneal-en-cot
120
+ pipeline_tag: text-generation
121
+ ---
122
+
123
+ # YuE-s1-7B-anneal-en-cot-exl2
124
+
125
+ [m-a-p/YuE-s1-7B-anneal-en-cot](https://huggingface.co/m-a-p/YuE-s1-7B-anneal-en-cot) quantized with Exllamav2. It appears to remain coherent using the default calibration data without adding audio tokens.
126
+
127
+ Intended to be used with the [WIP exl2 inference repository for YuE](https://github.com/sgsdxzy/YuE-exllamav2).
128
+
129
+ Links:
130
+ - [4.25bpw-h6](https://huggingface.co/Doctor-Shotgun/YuE-s1-7B-anneal-en-cot-exl2/tree/4.25bpw-h6)
131
+ - [5.0bpw-h6](https://huggingface.co/Doctor-Shotgun/YuE-s1-7B-anneal-en-cot-exl2/tree/5.0bpw-h6)
132
+ - [6.0bpw-h6](https://huggingface.co/Doctor-Shotgun/YuE-s1-7B-anneal-en-cot-exl2/tree/6.0bpw-h6)
133
+ - [8.0bpw-h8](https://huggingface.co/Doctor-Shotgun/YuE-s1-7B-anneal-en-cot-exl2/tree/8.0bpw-h8)","{""id"": ""Doctor-Shotgun/YuE-s1-7B-anneal-en-cot-exl2"", ""author"": ""Doctor-Shotgun"", ""sha"": ""f98177d5acaf8c48936f2ea470789cf40307f4d7"", ""last_modified"": ""2025-02-01 00:13:22+00:00"", ""created_at"": ""2025-01-31 23:14:12+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 93, ""downloads_all_time"": null, ""likes"": 10, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""music"", ""llama"", ""text-generation"", ""en"", ""base_model:m-a-p/YuE-s1-7B-anneal-en-cot"", ""base_model:finetune:m-a-p/YuE-s1-7B-anneal-en-cot"", ""license:apache-2.0"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: m-a-p/YuE-s1-7B-anneal-en-cot\nlanguage:\n- en\nlicense: apache-2.0\npipeline_tag: text-generation\ntags:\n- music\n- llama\nquantized_by: Doctor-Shotgun"", ""widget_data"": [{""text"": ""My name is Julien and I like to""}, {""text"": ""I like traveling by train because""}, {""text"": ""Paris is an amazing place to visit,""}, {""text"": ""Once upon a time,""}], ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='measurement.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-01 00:13:22+00:00"", ""cardData"": ""base_model: m-a-p/YuE-s1-7B-anneal-en-cot\nlanguage:\n- en\nlicense: apache-2.0\npipeline_tag: text-generation\ntags:\n- music\n- llama\nquantized_by: Doctor-Shotgun"", ""transformersInfo"": null, ""_id"": ""679d5944575df6520d55c900"", ""modelId"": ""Doctor-Shotgun/YuE-s1-7B-anneal-en-cot-exl2"", ""usedStorage"": 19191927672}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=Doctor-Shotgun/YuE-s1-7B-anneal-en-cot-exl2&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BDoctor-Shotgun%2FYuE-s1-7B-anneal-en-cot-exl2%5D(%2FDoctor-Shotgun%2FYuE-s1-7B-anneal-en-cot-exl2)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
adetailer_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv ADDED
@@ -0,0 +1,105 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ Bingsu/adetailer,"---
3
+ license: apache-2.0
4
+ library_name: ultralytics
5
+ datasets:
6
+ - wider_face
7
+ - skytnt/anime-segmentation
8
+ tags:
9
+ - pytorch
10
+ ---
11
+
12
+ # YOLOv8 Detection Model
13
+
14
+ ## Datasets
15
+
16
+ ### Face
17
+
18
+ - [Anime Face CreateML](https://universe.roboflow.com/my-workspace-mph8o/anime-face-createml)
19
+ - [xml2txt](https://universe.roboflow.com/0oooooo0/xml2txt-njqx1)
20
+ - [AN](https://universe.roboflow.com/sed-b8vkf/an-lfg5i)
21
+ - [wider face](http://shuoyang1213.me/WIDERFACE/index.html)
22
+
23
+ ### Hand
24
+
25
+ - [AnHDet](https://universe.roboflow.com/1-yshhi/anhdet)
26
+ - [hand-detection-fuao9](https://universe.roboflow.com/catwithawand/hand-detection-fuao9)
27
+
28
+ ### Person
29
+
30
+ - [coco2017](https://cocodataset.org/#home) (only person)
31
+ - [AniSeg](https://github.com/jerryli27/AniSeg)
32
+ - [skytnt/anime-segmentation](https://huggingface.co/datasets/skytnt/anime-segmentation)
33
+
34
+ ### deepfashion2
35
+
36
+ - [deepfashion2](https://github.com/switchablenorms/DeepFashion2)
37
+
38
+ | id | label |
39
+ | --- | --------------------- |
40
+ | 0 | short_sleeved_shirt |
41
+ | 1 | long_sleeved_shirt |
42
+ | 2 | short_sleeved_outwear |
43
+ | 3 | long_sleeved_outwear |
44
+ | 4 | vest |
45
+ | 5 | sling |
46
+ | 6 | shorts |
47
+ | 7 | trousers |
48
+ | 8 | skirt |
49
+ | 9 | short_sleeved_dress |
50
+ | 10 | long_sleeved_dress |
51
+ | 11 | vest_dress |
52
+ | 12 | sling_dress |
53
+
54
+ ## Info
55
+
56
+ | Model | Target | mAP 50 | mAP 50-95 |
57
+ | --------------------------- | --------------------- | ----------------------------- | ----------------------------- |
58
+ | face_yolov8n.pt | 2D / realistic face | 0.660 | 0.366 |
59
+ | face_yolov8n_v2.pt | 2D / realistic face | 0.669 | 0.372 |
60
+ | face_yolov8s.pt | 2D / realistic face | 0.713 | 0.404 |
61
+ | face_yolov8m.pt | 2D / realistic face | 0.737 | 0.424 |
62
+ | face_yolov9c.pt | 2D / realistic face | 0.748 | 0.433 |
63
+ | hand_yolov8n.pt | 2D / realistic hand | 0.767 | 0.505 |
64
+ | hand_yolov8s.pt | 2D / realistic hand | 0.794 | 0.527 |
65
+ | hand_yolov9c.pt | 2D / realistic hand | 0.810 | 0.550 |
66
+ | person_yolov8n-seg.pt | 2D / realistic person | 0.782 (bbox)<br/>0.761 (mask) | 0.555 (bbox)<br/>0.460 (mask) |
67
+ | person_yolov8s-seg.pt | 2D / realistic person | 0.824 (bbox)<br/>0.809 (mask) | 0.605 (bbox)<br/>0.508 (mask) |
68
+ | person_yolov8m-seg.pt | 2D / realistic person | 0.849 (bbox)<br/>0.831 (mask) | 0.636 (bbox)<br/>0.533 (mask) |
69
+ | deepfashion2_yolov8s-seg.pt | realistic clothes | 0.849 (bbox)<br/>0.840 (mask) | 0.763 (bbox)<br/>0.675 (mask) |
70
+
71
+ ## Usage
72
+
73
+ ```python
74
+ from huggingface_hub import hf_hub_download
75
+ from ultralytics import YOLO
76
+
77
+ path = hf_hub_download(""Bingsu/adetailer"", ""face_yolov8n.pt"")
78
+ model = YOLO(path)
79
+ ```
80
+
81
+ ```python
82
+ import cv2
83
+ from PIL import Image
84
+
85
+ img = ""https://farm5.staticflickr.com/4139/4887614566_6b57ec4422_z.jpg""
86
+ output = model(img)
87
+ pred = output[0].plot()
88
+ pred = cv2.cvtColor(pred, cv2.COLOR_BGR2RGB)
89
+ pred = Image.fromarray(pred)
90
+ pred
91
+ ```
92
+
93
+ ![image](https://i.imgur.com/9ny1wmD.png)
94
+
95
+
96
+ ## Unsafe files
97
+
98
+ ![image](https://i.imgur.com/9Btuy8j.png)
99
+
100
+ Since `getattr` is classified as a dangerous pickle function, any segmentation model that uses it is classified as unsafe.
101
+
102
+ All models were created and saved using the official [ultralytics](https://github.com/ultralytics/ultralytics) library, so it's okay to use files downloaded from a trusted source.
103
+
104
+ See also: https://huggingface.co/docs/hub/security-pickle
105
+ ","{""id"": ""Bingsu/adetailer"", ""author"": ""Bingsu"", ""sha"": ""53cc19de382014514d9d4038601d261a7faa9b7b"", ""last_modified"": ""2024-11-21 12:40:27+00:00"", ""created_at"": ""2023-04-26 00:58:45+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 21666424, ""downloads_all_time"": null, ""likes"": 569, ""library_name"": ""ultralytics"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""ultralytics"", ""pytorch"", ""dataset:wider_face"", ""dataset:skytnt/anime-segmentation"", ""doi:10.57967/hf/3633"", ""license:apache-2.0"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""datasets:\n- wider_face\n- skytnt/anime-segmentation\nlibrary_name: ultralytics\nlicense: apache-2.0\ntags:\n- pytorch"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='deepfashion2_yolov8s-seg.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='face_yolov8m.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='face_yolov8n.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='face_yolov8n_v2.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='face_yolov8s.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='face_yolov9c.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hand_yolov8n.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hand_yolov8s.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='hand_yolov9c.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='person_yolov8m-seg.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='person_yolov8n-seg.pt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='person_yolov8s-seg.pt', size=None, blob_id=None, lfs=None)""], ""spaces"": [""jhj0517/AdvancedLivePortrait-WebUI"", ""Aatricks/LightDiffusion-Next"", ""Jeffgold/adetailer"", ""hhxxhh/a1"", ""Bingsu/adtriton"", ""gartajackhats1985/custom_nodes""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-11-21 12:40:27+00:00"", ""cardData"": ""datasets:\n- wider_face\n- skytnt/anime-segmentation\nlibrary_name: ultralytics\nlicense: apache-2.0\ntags:\n- pytorch"", ""transformersInfo"": null, ""_id"": ""644877453e498d66919f36ec"", ""modelId"": ""Bingsu/adetailer"", ""usedStorage"": 1130245365}",0,,0,,0,,0,,0,"Aatricks/LightDiffusion-Next, Bingsu/adtriton, Jeffgold/adetailer, gartajackhats1985/custom_nodes, hhxxhh/a1, huggingface/InferenceSupport/discussions/new?title=Bingsu/adetailer&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BBingsu%2Fadetailer%5D(%2FBingsu%2Fadetailer)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, jhj0517/AdvancedLivePortrait-WebUI",7
bark-small_finetunes_20250426_215237.csv_finetunes_20250426_215237.csv ADDED
The diff for this file is too large to render. See raw diff
 
bert-base-NER_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv ADDED
The diff for this file is too large to render. See raw diff
 
chatglm-6b_finetunes_20250424_145241.csv_finetunes_20250424_145241.csv ADDED
@@ -0,0 +1,164 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ THUDM/chatglm-6b,"---
3
+ language:
4
+ - zh
5
+ - en
6
+ tags:
7
+ - glm
8
+ - chatglm
9
+ - thudm
10
+ ---
11
+ # ChatGLM-6B
12
+ <p align=""center"">
13
+ 🌐 <a href=""https://chatglm.cn/blog"" target=""_blank"">Blog</a> • 💻 <a href=""https://github.com/THUDM/ChatGLM-6B"" target=""_blank"">Github Repo</a> • 🐦 <a href=""https://twitter.com/thukeg"" target=""_blank"">Twitter</a> • 📃 <a href=""https://arxiv.org/abs/2103.10360"" target=""_blank"">[GLM@ACL 22]</a> <a href=""https://github.com/THUDM/GLM"" target=""_blank"">[GitHub]</a> • 📃 <a href=""https://arxiv.org/abs/2210.02414"" target=""_blank"">[GLM-130B@ICLR 23]</a> <a href=""https://github.com/THUDM/GLM-130B"" target=""_blank"">[GitHub]</a> <br>
14
+ </p>
15
+
16
+ <p align=""center"">
17
+ 👋 Join our <a href=""https://join.slack.com/t/chatglm/shared_invite/zt-1y7pqoloy-9b1g6T6JjA8J0KxvUjbwJw"" target=""_blank"">Slack</a> and <a href=""https://github.com/THUDM/ChatGLM-6B/blob/main/resources/WECHAT.md"" target=""_blank"">WeChat</a>
18
+ </p>
19
+
20
+ <p align=""center"">
21
+ 📍Experience the larger-scale ChatGLM model at <a href=""https://www.chatglm.cn"">chatglm.cn</a>
22
+ </p>
23
+
24
+ **我们发布了 [ChatGLM2-6B](https://github.com/THUDM/ChatGLM2-6B)，ChatGLM-6B 的升级版本，在保留了了初代模型对话流畅、部署门槛较低等众多优秀特性的基础之上，引入了更强大的性能、更长的上下文、更高效的推理等升级。**
25
+ ## 介绍
26
+ ChatGLM-6B 是一个开源的、支持中英双语问答的对话语言模型，基于 [General Language Model (GLM)](https://github.com/THUDM/GLM) 架构，具有 62 亿参数。结合模型量化技术，用户可以在消费级的显卡上进行本地部署（INT4 量化级别下最低只需 6GB 显存）。ChatGLM-6B 使用了和 [ChatGLM](https://chatglm.cn) 相同的技术，针对中文问答和对话进行了优化。经过约 1T 标识符的中英双语训练，辅以监督微调、反馈自助、人类反馈强化学习等技术的加持，62 亿参数的 ChatGLM-6B 已经能生成相当符合人类偏好的回答。 ChatGLM-6B 权重对学术研究**完全开放**，在填写[问卷](https://open.bigmodel.cn/mla/form)进行登记后**亦允许免费商业使用**。
27
+
28
+ ChatGLM-6B is an open bilingual language model based on [General Language Model (GLM)](https://github.com/THUDM/GLM) framework, with 6.2 billion parameters. With the quantization technique, users can deploy locally on consumer-grade graphics cards (only 6GB of GPU memory is required at the INT4 quantization level). ChatGLM-6B uses technology similar to ChatGPT, optimized for Chinese QA and dialogue. The model is trained for about 1T tokens of Chinese and English corpus, supplemented by supervised fine-tuning, feedback bootstrap, and reinforcement learning with human feedback. With only about 6.2 billion parameters, the model is able to generate answers that are in line with human preference. ChatGLM-6B weights are **completely open** for academic research, and **free commercial use** is also allowed after completing the [questionnaire](https://open.bigmodel.cn/mla/form).
29
+
30
+ ## 软件依赖
31
+
32
+ ```shell
33
+ pip install protobuf==3.20.0 transformers==4.27.1 icetk cpm_kernels
34
+ ```
35
+
36
+ ## 代码调用
37
+
38
+ 可以通过如下代码调用 ChatGLM-6B 模型来生成对话：
39
+
40
+ ```ipython
41
+ >>> from transformers import AutoTokenizer, AutoModel
42
+ >>> tokenizer = AutoTokenizer.from_pretrained(""THUDM/chatglm-6b"", trust_remote_code=True)
43
+ >>> model = AutoModel.from_pretrained(""THUDM/chatglm-6b"", trust_remote_code=True).half().cuda()
44
+ >>> response, history = model.chat(tokenizer, ""你好"", history=[])
45
+ >>> print(response)
46
+ 你好👋!我是人工智能助手 ChatGLM-6B,很高兴见到你,欢迎问我任何问题。
47
+ >>> response, history = model.chat(tokenizer, ""晚上睡不着应该怎么办"", history=history)
48
+ >>> print(response)
49
+ 晚上睡不着可能会让你感到焦虑或不舒服,但以下是一些可以帮助你入睡的方法:
50
+
51
+ 1. 制定规律的睡眠时间表:保持规律的睡眠时间表可以帮助你建立健康的睡眠习惯,使你更容易入睡。尽量在每天的相同时间上床,并在同一时间起床。
52
+ 2. 创造一个舒适的睡眠环境:确保睡眠环境舒适,安静,黑暗且温度适宜。可以使用舒适的床上用品,并保持房间通风。
53
+ 3. 放松身心:在睡前做些放松的活动,例如泡个热水澡,听些轻柔的音乐,阅读一些有趣的书籍等,有助于缓解紧张和焦虑,使你更容易入睡。
54
+ 4. 避免饮用含有咖啡因的饮料:咖啡因是一种刺激性物质,会影响你的睡眠质量。尽量避免在睡前饮用含有咖啡因的饮料,例如咖啡,茶和可乐。
55
+ 5. 避免在床上做与睡眠无关的事情:在床上做些与睡眠无关的事情,例如看电影,玩游戏或工作等,可能会干扰你的睡眠。
56
+ 6. 尝试呼吸技巧:深呼吸是一种放松技巧,可以帮助你缓解紧张和焦虑,使你更容易入睡。试着���慢吸气,保持几秒钟,然后缓慢呼气。
57
+
58
+ 如果这些方法无法帮助你入睡,你可以考虑咨询医生或睡眠专家,寻求进一步的建议。
59
+ ```
60
+
61
+ 关于更多的使用说明，包括如何运行命令行和网页版本的 DEMO，以及使用模型量化以节省显存，请参考我们的 [Github Repo](https://github.com/THUDM/ChatGLM-6B)。
62
+
63
+ For more instructions, including how to run CLI and web demos, and model quantization, please refer to our [Github Repo](https://github.com/THUDM/ChatGLM-6B).
64
+
65
+ ## Change Log
66
+ * v1.1.0 ([942945d](https://huggingface.co/THUDM/chatglm-6b/commit/942945df047dee66f653c68ae0e56655045f1741)): 更新 v1.1 版本 checkpoint
67
+ * v0.1.0 ([f831824](https://huggingface.co/THUDM/chatglm-6b/commit/f83182484538e663a03d3f73647f10f89878f438))
68
+
69
+ ## 协议
70
+
71
+ 本仓库的代码依照 [Apache-2.0](LICENSE) 协议开源，ChatGLM-6B 模型的权重的使用则需要遵循 [Model License](MODEL_LICENSE)。
72
+
73
+ ## 引用
74
+
75
+ 如果你觉得我们的工作有帮助的话，请考虑引用下列论文。
76
+
77
+ If you find our work helpful, please consider citing the following paper.
78
+
79
+ ```
80
+ @misc{glm2024chatglm,
81
+ title={ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools},
82
+ author={Team GLM and Aohan Zeng and Bin Xu and Bowen Wang and Chenhui Zhang and Da Yin and Diego Rojas and Guanyu Feng and Hanlin Zhao and Hanyu Lai and Hao Yu and Hongning Wang and Jiadai Sun and Jiajie Zhang and Jiale Cheng and Jiayi Gui and Jie Tang and Jing Zhang and Juanzi Li and Lei Zhao and Lindong Wu and Lucen Zhong and Mingdao Liu and Minlie Huang and Peng Zhang and Qinkai Zheng and Rui Lu and Shuaiqi Duan and Shudan Zhang and Shulin Cao and Shuxun Yang and Weng Lam Tam and Wenyi Zhao and Xiao Liu and Xiao Xia and Xiaohan Zhang and Xiaotao Gu and Xin Lv and Xinghan Liu and Xinyi Liu and Xinyue Yang and Xixuan Song and Xunkai Zhang and Yifan An and Yifan Xu and Yilin Niu and Yuantao Yang and Yueyan Li and Yushi Bai and Yuxiao Dong and Zehan Qi and Zhaoyu Wang and Zhen Yang and Zhengxiao Du and Zhenyu Hou and Zihan Wang},
83
+ year={2024},
84
+ eprint={2406.12793},
85
+ archivePrefix={arXiv},
86
+ primaryClass={id='cs.CL' full_name='Computation and Language' is_active=True alt_name='cmp-lg' in_archive='cs' is_general=False description='Covers natural language processing. Roughly includes material in ACM Subject Class I.2.7. Note that work on artificial languages (programming languages, logics, formal systems) that does not explicitly address natural-language issues broadly construed (natural-language processing, computational linguistics, speech, text retrieval, etc.) is not appropriate for this area.'}
87
+ }
88
+ ```","{""id"": ""THUDM/chatglm-6b"", ""author"": ""THUDM"", ""sha"": ""bf0f5cfb575eebebf9b655c5861177acfee03f16"", ""last_modified"": ""2024-08-04 08:44:58+00:00"", ""created_at"": ""2023-03-13 16:28:04+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 4145, ""downloads_all_time"": null, ""likes"": 2854, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""pytorch"", ""chatglm"", ""glm"", ""thudm"", ""custom_code"", ""zh"", ""en"", ""arxiv:2103.10360"", ""arxiv:2210.02414"", ""arxiv:2406.12793"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""language:\n- zh\n- en\ntags:\n- glm\n- chatglm\n- thudm"", ""widget_data"": null, ""model_index"": null, ""config"": {""architectures"": [""ChatGLMModel""], ""auto_map"": {""AutoConfig"": ""configuration_chatglm.ChatGLMConfig"", ""AutoModel"": ""modeling_chatglm.ChatGLMForConditionalGeneration"", ""AutoModelForSeq2SeqLM"": ""modeling_chatglm.ChatGLMForConditionalGeneration""}, ""model_type"": ""chatglm"", ""tokenizer_config"": {""bos_token"": ""<sop>"", ""eos_token"": ""<eop>"", ""mask_token"": ""[MASK]"", ""pad_token"": ""<pad>"", ""unk_token"": ""<unk>""}}, ""transformers_info"": {""auto_model"": ""AutoModel"", ""custom_class"": null, ""pipeline_tag"": null, ""processor"": null}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='LICENSE', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='MODEL_LICENSE', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='configuration_chatglm.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ice_text.model', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='modeling_chatglm.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00001-of-00008.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00002-of-00008.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00003-of-00008.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00004-of-00008.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00005-of-00008.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00006-of-00008.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00007-of-00008.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00008-of-00008.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model.bin.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='quantization.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='test_modeling_chatglm.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenization_chatglm.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [""multimodalart/ChatGLM-6B"", ""ChallengeHub/Chinese-LangChain"", ""OptimalScale/Robin-7b"", ""ysharma/ChatGLM-6b_Gradio_Streaming"", ""John6666/joy-caption-pre-alpha-mod"", ""aimevzulari/Prompt_Uzmani"", ""DrSong/ChatGLM-6B-ChatBot"", ""ysharma/OSChatbots_ChatGPT_ToeToToe"", ""josStorer/ChatGLM-6B-Int4-API-OpenAI-Compatible"", ""meval/multilingual-chatbot-arena-leaderboard"", ""EmbeddedLLM/chat-template-generation"", ""qingxu98/academic-chatgpt-beta"", ""llm-blender/LLM-Blender"", ""hahahafofo/ChatGLM-Chinese-Summary"", ""ljsabc/Fujisaki"", ""yhavinga/dutch-tokenizer-arena"", ""Kevin676/Shanghainese-TTS-demo"", ""suchun/chatGPT_acdemic"", ""linxianzhong0128/Linly-Talker"", ""hahahafofo/ChatPDF"", ""Ameaou/academic-chatgpt3.1"", ""li-qing/FIRE"", ""fkhuggingme/gpt-academic"", ""alitrack/ChatPDF"", ""OedoSoldier/chatglm_int4_demo"", ""Dao3/ChatGLM-6B"", ""Yiqin/ChatVID"", ""elitecode/ChatGLM-6B-ChatBot"", ""OptimalScale/Robin-33b"", ""f2api/gpt-academic"", ""syx948/ChatPDF"", ""CodeTed/Chinese-Grammarly"", ""CodeTed/chinese_spelling_error_correction"", ""MegaTronX/joy-caption-pre-alpha-mod"", ""TongkunGuan/Token-level_Text_Image_Foundation_Model"", ""fb700/chat3"", ""Gmq-x/gpt-academic"", ""IntelligenzaArtificiale/ChatGLM-6B-Int4-API-OpenAI-Compatible"", ""Kevin676/Telephone-Interviewing_PpaddleSpeech-TTS"", ""YukiKurosawaDev/ChatGLM"", ""kevinwang676/ChatGLM-int4-demo"", ""erbanku/gpt-academic"", ""caojiachen1/chatgpt-webui"", ""Fengbinbin/gpt-academic"", ""xwsm/gpt"", ""Cong723/gpt-academic-public"", ""weiwei1392/paper_generate"", ""bigPear/digitalWDF"", ""aodianyun/ChatGLM-6B"", ""hbestm/gpt-academic-play"", ""actboy/ChatGLM-6B"", ""xxccc/gpt-academic"", ""sinksmell/ChatPDF"", ""dakaiye/dky_xuexi"", ""hands012/gpt-academic"", ""shawn810720/Taiwan-LLaMa2"", ""tianleliphoebe/visual-arena"", ""Ashmal/MobiLlama"", ""JIAFENG7/BFF"", ""Nickofranco/Sigma"", ""Nerva5678/Excel-QA-bot"", ""Csdfg/ChatGLM-6B-Int4-API-OpenAI-Compatible"", ""Ligeng-Zhu/Digital-Clone"", ""oscarmei/oscar"", ""henryu/ChatGLB_007"", ""cndavy/ChatGLM-6B"", ""zefanwang/gpt-academic"", ""lockinwu/lockinwu-ChatGLM-6B"", ""fariliang/ChatGLM-Batman"", ""zedchou/gpt-academic"", ""KumaTea/KumaGLM"", ""slamkkk/chatgpt_academic"", ""ranv5/open"", ""marvin68/chatglm6b"", ""Here123/play"", ""KumaTea/KumaGLM-Lite"", ""gaoshine/Chinese-LangChain"", ""saintjohn/gpt-academicpub"", ""long1111/chatglm"", ""chad2020/gptzhongkeyuan"", ""roman008/gpt-academic"", ""HMinions/ws"", ""NewBreaker/chatglm-6b-int4"", ""kevinwang676/ChatPDF"", ""ydm8964/jjygpt"", ""Danielzero/gpt-academic"", ""ycy1615/gpt-academic"", ""leemingzhe/gpt-academic-zafu-7-301"", ""randomchaos9999/xsrwascq"", ""acedemic-talent/gpt-academic"", ""tree1024/gpt-academic"", ""zdevc/chatai"", ""Elphen/gpt-academic3.32"", ""pgqun/gpt-academic"", ""jackli888/chatGPT_acdemic"", ""zhang55/gpt-academic-zafu-7-301"", ""cgisky/gpt-academic22"", ""Jasonai/gpt-academic"", ""AtonB/gpt-academic"", ""MoDongbao/ChatGPT_for_Academic_Releases96""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2024-08-04 08:44:58+00:00"", ""cardData"": ""language:\n- zh\n- en\ntags:\n- glm\n- chatglm\n- thudm"", ""transformersInfo"": {""auto_model"": ""AutoModel"", ""custom_class"": null, ""pipeline_tag"": null, ""processor"": null}, ""_id"": ""640f4f1409c94e1d9bca3ffc"", ""modelId"": ""THUDM/chatglm-6b"", ""usedStorage"": 47452655921}",0,https://huggingface.co/Longz1003/chatglm_lora_mimeng,1,https://huggingface.co/shibing624/chatglm-6b-belle-zh-lora,1,,0,,0,"Ashmal/MobiLlama, CodeTed/Chinese-Grammarly, CodeTed/chinese_spelling_error_correction, Fengbinbin/gpt-academic, John6666/joy-caption-pre-alpha-mod, MegaTronX/joy-caption-pre-alpha-mod, TongkunGuan/Token-level_Text_Image_Foundation_Model, aimevzulari/Prompt_Uzmani, huggingface/InferenceSupport/discussions/new?title=THUDM/chatglm-6b&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BTHUDM%2Fchatglm-6b%5D(%2FTHUDM%2Fchatglm-6b)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, linxianzhong0128/Linly-Talker, meval/multilingual-chatbot-arena-leaderboard, qingxu98/academic-chatgpt-beta, yhavinga/dutch-tokenizer-arena",13
89
+ Longz1003/chatglm_lora_mimeng,"---
90
+ license: apache-2.0
91
+ datasets:
92
+ - fka/awesome-chatgpt-prompts
93
+ language:
94
+ - zh
95
+ base_model:
96
+ - THUDM/chatglm-6b
97
+ pipeline_tag: question-answering
98
+ ---
99
+
100
+ # ChatGLM-3-6B-LoRA: A LoRA Fine-Tuned Version of ChatGLM-3-6B
101
+
102
+ ## Model description
103
+ 这是一个基于 **ChatGLM-3-6B** 进行 **LoRA 微调** 的中文对话生成模型。该模型的目标是提升对话生成的质量，特别是在开放式对话和问答任务中。LoRA 微调技术通过低秩适配层优化了预训练模型，减少了计算开销。
104
+
105
+ 目前，模型在对话生成的质量上会继续进行微调和改进，以优化性能和适应更多场景。
106
+
107
+ ## Model details
108
+ - **架构**：基于 **ChatGLM-3-6B**（一个大规模中文生成模型）。
109
+ - **微调技术**：使用 **LoRA** 微调，对预训练的 ChatGLM 模型进行低秩调整。
110
+ - **任务类型**：主要用于中文对话生成、问答任务等。
111
+
112
+ ## Training data
113
+ - **训练数据集**：使用自定义对话数据集进行微调。
114
+ - **数据量和多样性**：目前的训练数据集质量有待提升，可能影响模型的泛化能力。
115
+ - **清洗与处理**：数据经过一定程度的清洗，去除噪声和低质量对话，但由于数据集的限制，某些任务的表现仍然较弱。
116
+
117
+ ## Evaluation results
118
+ 模型在一些对话生成任务上的初步评估结果如下：
119
+ - **准确性**：当前在常见对话和开放式问答任务中的表现较好。
120
+ - **流畅度**：生成的对话大部分时间较为流畅，但偶尔出现重复、无关或不自然的回答。
121
+ - **性能瓶颈**：由于训练数据的局限性，模型在长时间对话中可能会出现表现下降的情况。
122
+
123
+ **当前模型的表现并未达到最佳效果，但随着更多数据和进一步的微调，预期会有所改进。**
124
+
125
+ ## Limitations and Biases
126
+ - **生成质量**：当前模型在复杂对话中的表现较为稳定，特别是在涉及特定领域或长时间对话时，生成的内容可能不准确或缺乏逻辑性。
127
+ - **数据偏差**：由于使用的训练数据来自公开数据集，模型可能在特定领域（如医学、科技等）表现较差，也可能存在一些基于数据集的偏见。
128
+ - **更新和优化**：模型仍在不断微调和优化中，效果可能会随着时间和进一步训练得到改善。
129
+
130
+ ## Use case
131
+ 这个模型适用于构建中文对话系统，特别是在：
132
+ - **聊天机器人**：构建能够进行简单对话和闲聊的聊天机器人。
133
+ - **教育辅助**：作为教育领域的智能问答助手，帮助学生解答基础问题。
134
+
135
+ ## How to use
136
+ 你可以使用以下代码来加载并使用该模型：
137
+
138
+ ```python
139
+ from transformers import AutoModelForCausalLM, AutoTokenizer
140
+
141
+ # 加载模型和 tokenizer
142
+ model_name = ""Longz1003/chatglm_lora_mimeng"" # 替换为你的模型路径
143
+ model = AutoModelForCausalLM.from_pretrained(model_name)
144
+ tokenizer = AutoTokenizer.from_pretrained(model_name)
145
+
146
+ # 测试模型
147
+ inputs = tokenizer(""你好，今天的天气怎么样？"", return_tensors=""pt"")
148
+ outputs = model.generate(**inputs)
149
+ print(tokenizer.decode(outputs[0], skip_special_tokens=True))
150
+
151
+ torch>=1.10.0
152
+ transformers>=4.30.0
153
+ accelerate>=0.18.0
154
+ datasets>=2.0.0
155
+ scipy>=1.7.0
156
+ sentencepiece>=0.1.96
157
+ pytorch-lightning>=1.5.0
158
+ pyyaml>=5.4.1
159
+ tqdm>=4.62.0
160
+ requests>=2.26.0
161
+ huggingface_hub>=0.12.0
162
+ tensorboard>=2.6.0
163
+ scikit-learn>=0.24.2
164
+ ","{""id"": ""Longz1003/chatglm_lora_mimeng"", ""author"": ""Longz1003"", ""sha"": ""6aca4e3a996be80815456c56ce54a53289fa95ed"", ""last_modified"": ""2025-04-08 03:44:12+00:00"", ""created_at"": ""2025-02-07 03:28:26+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 9, ""downloads_all_time"": null, ""likes"": 1, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""safetensors"", ""chatglm"", ""question-answering"", ""custom_code"", ""zh"", ""dataset:fka/awesome-chatgpt-prompts"", ""base_model:THUDM/chatglm-6b"", ""base_model:finetune:THUDM/chatglm-6b"", ""license:apache-2.0"", ""region:us""], ""pipeline_tag"": ""question-answering"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model:\n- THUDM/chatglm-6b\ndatasets:\n- fka/awesome-chatgpt-prompts\nlanguage:\n- zh\nlicense: apache-2.0\npipeline_tag: question-answering"", ""widget_data"": [{""text"": ""\u6211\u4f4f\u5728\u54ea\u91cc\uff1f"", ""context"": ""\u6211\u53eb\u6c83\u5c14\u592b\u5188\uff0c\u6211\u4f4f\u5728\u67cf\u6797\u3002""}, {""text"": ""\u6211\u4f4f\u5728\u54ea\u91cc\uff1f"", ""context"": ""\u6211\u53eb\u8428\u62c9\uff0c\u6211\u4f4f\u5728\u4f26\u6566\u3002""}, {""text"": ""\u6211\u7684\u540d\u5b57\u662f\u4ec0\u4e48\uff1f"", ""context"": ""\u6211\u53eb\u514b\u62c9\u62c9\uff0c\u6211\u4f4f\u5728\u4f2f\u514b\u5229\u3002""}], ""model_index"": null, ""config"": {""architectures"": [""ChatGLMForConditionalGeneration""], ""auto_map"": {""AutoConfig"": ""configuration_chatglm.ChatGLMConfig"", ""AutoModel"": ""modeling_chatglm.ChatGLMForConditionalGeneration"", ""AutoModelForCausalLM"": ""modeling_chatglm.ChatGLMForConditionalGeneration"", ""AutoModelForSeq2SeqLM"": ""modeling_chatglm.ChatGLMForConditionalGeneration"", ""AutoModelForSequenceClassification"": ""modeling_chatglm.ChatGLMForSequenceClassification""}, ""model_type"": ""chatglm"", ""tokenizer_config"": {""chat_template"": ""{% for message in messages %}{% if loop.first %}[gMASK]sop<|{{ message['role'] }}|>\n {{ message['content'] }}{% else %}<|{{ message['role'] }}|>\n {{ message['content'] }}{% endif %}{% endfor %}{% if add_generation_prompt %}<|assistant|>{% endif %}"", ""eos_token"": ""</s>"", ""pad_token"": ""<unk>"", ""unk_token"": ""<unk>""}}, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='configuration_chatglm.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00001-of-00007.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00002-of-00007.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00003-of-00007.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00004-of-00007.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00005-of-00007.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00006-of-00007.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model-00007-of-00007.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='model.safetensors.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='modeling_chatglm.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='quantization.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenization_chatglm.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.model', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": {""parameters"": {""F16"": 6243584032}, ""total"": 6243584032}, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-04-08 03:44:12+00:00"", ""cardData"": ""base_model:\n- THUDM/chatglm-6b\ndatasets:\n- fka/awesome-chatgpt-prompts\nlanguage:\n- zh\nlicense: apache-2.0\npipeline_tag: question-answering"", ""transformersInfo"": null, ""_id"": ""67a57ddacdd8bd444049b210"", ""modelId"": ""Longz1003/chatglm_lora_mimeng"", ""usedStorage"": 12488213274}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=Longz1003/chatglm_lora_mimeng&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BLongz1003%2Fchatglm_lora_mimeng%5D(%2FLongz1003%2Fchatglm_lora_mimeng)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
codebert-base_finetunes_20250426_171734.csv_finetunes_20250426_171734.csv ADDED
The diff for this file is too large to render. See raw diff
 
colpali_finetunes_20250425_165642.csv_finetunes_20250425_165642.csv ADDED
@@ -0,0 +1,154 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ vidore/colpali,"---
3
+ license: mit
4
+ library_name: colpali
5
+ base_model: google/paligemma-3b-mix-448
6
+ language:
7
+ - en
8
+ tags:
9
+ - colpali
10
+ - vidore
11
+ new_version: vidore/colpali-v1.1
12
+ datasets:
13
+ - vidore/colpali_train_set
14
+ pipeline_tag: visual-document-retrieval
15
+ ---
16
+ # ColPali: Visual Retriever based on PaliGemma-3B with ColBERT strategy
17
+
18
+ ColPali is a model based on a novel model architecture and training strategy based on Vision Language Models (VLMs) to efficiently index documents from their visual features.
19
+ It is a [PaliGemma-3B](https://huggingface.co/google/paligemma-3b-mix-448) extension that generates [ColBERT](https://arxiv.org/abs/2004.12832)- style multi-vector representations of text and images.
20
+ It was introduced in the paper [ColPali: Efficient Document Retrieval with Vision Language Models](https://arxiv.org/abs/2407.01449) and first released in [this repository](https://github.com/ManuelFay/colpali)
21
+
22
+ <p align=""center""><img width=800 src=""https://github.com/illuin-tech/colpali/blob/main/assets/colpali_architecture.webp?raw=true""/></p>
23
+
24
+ ## Model Description
25
+
26
+ This model is built iteratively starting from an off-the-shelf [SigLIP](https://huggingface.co/google/siglip-so400m-patch14-384) model.
27
+ We finetuned it to create [BiSigLIP](https://huggingface.co/vidore/bisiglip) and fed the patch-embeddings output by SigLIP to an LLM, [PaliGemma-3B](https://huggingface.co/google/paligemma-3b-mix-448) to create [BiPali](https://huggingface.co/vidore/bipali).
28
+
29
+ One benefit of inputting image patch embeddings through a language model is that they are natively mapped to a latent space similar to textual input (query).
30
+ This enables leveraging the [ColBERT](https://arxiv.org/abs/2004.12832) strategy to compute interactions between text tokens and image patches, which enables a step-change improvement in performance compared to BiPali.
31
+
32
+ ## Model Training
33
+
34
+ ### Dataset
35
+ Our training dataset of 127,460 query-page pairs is comprised of train sets of openly available academic datasets (63%) and a synthetic dataset made up of pages from web-crawled PDF documents and augmented with VLM-generated (Claude-3 Sonnet) pseudo-questions (37%).
36
+ Our training set is fully English by design, enabling us to study zero-shot generalization to non-English languages. We explicitly verify no multi-page PDF document is used both [*ViDoRe*](https://huggingface.co/collections/vidore/vidore-benchmark-667173f98e70a1c0fa4db00d) and in the train set to prevent evaluation contamination.
37
+ A validation set is created with 2% of the samples to tune hyperparameters.
38
+
39
+ *Note: Multilingual data is present in the pretraining corpus of the language model (Gemma-2B) and potentially occurs during PaliGemma-3B's multimodal training.*
40
+
41
+ ### Parameters
42
+
43
+ All models are trained for 1 epoch on the train set. Unless specified otherwise, we train models in `bfloat16` format, use low-rank adapters ([LoRA](https://arxiv.org/abs/2106.09685))
44
+ with `alpha=32` and `r=32` on the transformer layers from the language model,
45
+ as well as the final randomly initialized projection layer, and use a `paged_adamw_8bit` optimizer.
46
+ We train on an 8 GPU setup with data parallelism, a learning rate of 5e-5 with linear decay with 2.5% warmup steps, and a batch size of 32.
47
+
48
+ ## Usage
49
+
50
+ ### For best performance, newer models are available (vidore/colpali-v1.2)
51
+
52
+
53
+ ```bash
54
+ # This model checkpoint is compatible with version 0.1.1, but not more recent versions of the inference lib
55
+ pip install colpali_engine==0.1.1
56
+ ```
57
+
58
+ ```python
59
+ import torch
60
+ import typer
61
+ from torch.utils.data import DataLoader
62
+ from tqdm import tqdm
63
+ from transformers import AutoProcessor
64
+ from PIL import Image
65
+
66
+ from colpali_engine.models.paligemma_colbert_architecture import ColPali
67
+ from colpali_engine.trainer.retrieval_evaluator import CustomEvaluator
68
+ from colpali_engine.utils.colpali_processing_utils import process_images, process_queries
69
+ from colpali_engine.utils.image_from_page_utils import load_from_dataset
70
+
71
+
72
+ def main() -> None:
73
+ """"""Example script to run inference with ColPali""""""
74
+
75
+ # Load model
76
+ model_name = ""vidore/colpali""
77
+ model = ColPali.from_pretrained(""vidore/colpaligemma-3b-mix-448-base"", torch_dtype=torch.bfloat16, device_map=""cuda"").eval()
78
+ model.load_adapter(model_name)
79
+ processor = AutoProcessor.from_pretrained(model_name)
80
+
81
+ # select images -> load_from_pdf(<pdf_path>), load_from_image_urls([""<url_1>""]), load_from_dataset(<path>)
82
+ images = load_from_dataset(""vidore/docvqa_test_subsampled"")
83
+ queries = [""From which university does James V. Fiorca come ?"", ""Who is the japanese prime minister?""]
84
+
85
+ # run inference - docs
86
+ dataloader = DataLoader(
87
+ images,
88
+ batch_size=4,
89
+ shuffle=False,
90
+ collate_fn=lambda x: process_images(processor, x),
91
+ )
92
+ ds = []
93
+ for batch_doc in tqdm(dataloader):
94
+ with torch.no_grad():
95
+ batch_doc = {k: v.to(model.device) for k, v in batch_doc.items()}
96
+ embeddings_doc = model(**batch_doc)
97
+ ds.extend(list(torch.unbind(embeddings_doc.to(""cpu""))))
98
+
99
+ # run inference - queries
100
+ dataloader = DataLoader(
101
+ queries,
102
+ batch_size=4,
103
+ shuffle=False,
104
+ collate_fn=lambda x: process_queries(processor, x, Image.new(""RGB"", (448, 448), (255, 255, 255))),
105
+ )
106
+
107
+ qs = []
108
+ for batch_query in dataloader:
109
+ with torch.no_grad():
110
+ batch_query = {k: v.to(model.device) for k, v in batch_query.items()}
111
+ embeddings_query = model(**batch_query)
112
+ qs.extend(list(torch.unbind(embeddings_query.to(""cpu""))))
113
+
114
+ # run evaluation
115
+ retriever_evaluator = CustomEvaluator(is_multi_vector=True)
116
+ scores = retriever_evaluator.evaluate(qs, ds)
117
+ print(scores.argmax(axis=1))
118
+
119
+
120
+ if __name__ == ""__main__"":
121
+ typer.run(main)
122
+
123
+ ```
124
+
125
+ ## Limitations
126
+
127
+ - **Focus**: The model primarily focuses on PDF-type documents and high-ressources languages, potentially limiting its generalization to other document types or less represented languages.
128
+ - **Support**: The model relies on multi-vector retreiving derived from the ColBERT late interaction mechanism, which may require engineering efforts to adapt to widely used vector retrieval frameworks that lack native multi-vector support.
129
+
130
+ ## License
131
+
132
+ ColPali's vision language backbone model (PaliGemma) is under `gemma` license as specified in its [model card](https://huggingface.co/google/paligemma-3b-mix-448). The adapters attached to the model are under MIT license.
133
+
134
+ ## Contact
135
+
136
+ - Manuel Faysse: manuel.faysse@illuin.tech
137
+ - Hugues Sibille: hugues.sibille@illuin.tech
138
+ - Tony Wu: tony.wu@illuin.tech
139
+
140
+ ## Citation
141
+
142
+ If you use any datasets or models from this organization in your research, please cite the original dataset as follows:
143
+
144
+ ```bibtex
145
+ @misc{faysse2024colpaliefficientdocumentretrieval,
146
+ title={ColPali: Efficient Document Retrieval with Vision Language Models},
147
+ author={Manuel Faysse and Hugues Sibille and Tony Wu and Bilel Omrani and Gautier Viaud and Céline Hudelot and Pierre Colombo},
148
+ year={2024},
149
+ eprint={2407.01449},
150
+ archivePrefix={arXiv},
151
+ primaryClass={cs.IR},
152
+ url={https://arxiv.org/abs/2407.01449},
153
+ }
154
+ ```","{""id"": ""vidore/colpali"", ""author"": ""vidore"", ""sha"": ""82b73faebcbdd2f2009ada076b22fb85d3b63665"", ""last_modified"": ""2025-02-05 19:23:14+00:00"", ""created_at"": ""2024-06-25 10:06:08+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 13831, ""downloads_all_time"": null, ""likes"": 436, ""library_name"": ""colpali"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""colpali"", ""safetensors"", ""vidore"", ""visual-document-retrieval"", ""en"", ""dataset:vidore/colpali_train_set"", ""arxiv:2004.12832"", ""arxiv:2407.01449"", ""arxiv:2106.09685"", ""base_model:google/paligemma-3b-mix-448"", ""base_model:finetune:google/paligemma-3b-mix-448"", ""license:mit"", ""region:us""], ""pipeline_tag"": ""visual-document-retrieval"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: google/paligemma-3b-mix-448\ndatasets:\n- vidore/colpali_train_set\nlanguage:\n- en\nlibrary_name: colpali\nlicense: mit\npipeline_tag: visual-document-retrieval\ntags:\n- colpali\n- vidore\nnew_version: vidore/colpali-v1.1"", ""widget_data"": null, ""model_index"": null, ""config"": {""tokenizer_config"": {""bos_token"": ""<bos>"", ""eos_token"": ""<eos>"", ""pad_token"": ""<pad>"", ""unk_token"": ""<unk>"", ""use_default_system_prompt"": false}}, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='added_tokens.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='git_hash.txt', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='preprocessor_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='results.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.model', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='training_config.yml', size=None, blob_id=None, lfs=None)""], ""spaces"": [""merve/vision_papers"", ""DeF0017/OCR-using-Qwen2-VL"", ""Swekerr/Qwen2VL-OCR"", ""clayton07/qwen2-colpali-ocr"", ""gabrielaltay/vlmqa"", ""profchaos/OCR-APP"", ""gauri-sharan/test-two"", ""HUANG-Stephanie/cvquest-colpali"", ""lukiod/streamlit_qwen"", ""lukiod/test2"", ""lukiod/dock2"", ""Rick7799/Ocr1"", ""Pranathi1/Qwen2vl_RAG"", ""rk404/ocr_hi_en"", ""RufusRubin777/Qwen2VL-OCR_CPU"", ""Manogari2003/OCR-with-Keyword-Search-from-Image"", ""mannywho/webocr"", ""anvi27/ocr"", ""Yassmen/OCR_App"", ""Niha14/images-texts"", ""ayushb03/colpali-qwen2-ocr-search"", ""intuitive262/Doc_Reader"", ""gkaur13/ocrHindiEnglish"", ""gkaur13/hindiEnglishOCR"", ""Pulkit-28/OCR"", ""velaia/vision_papers""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-02-05 19:23:14+00:00"", ""cardData"": ""base_model: google/paligemma-3b-mix-448\ndatasets:\n- vidore/colpali_train_set\nlanguage:\n- en\nlibrary_name: colpali\nlicense: mit\npipeline_tag: visual-document-retrieval\ntags:\n- colpali\n- vidore\nnew_version: vidore/colpali-v1.1"", ""transformersInfo"": null, ""_id"": ""667a96900df5089affb1ba53"", ""modelId"": ""vidore/colpali"", ""usedStorage"": 100652593}",0,,0,,0,,0,,0,"DeF0017/OCR-using-Qwen2-VL, HUANG-Stephanie/cvquest-colpali, Rick7799/Ocr1, Swekerr/Qwen2VL-OCR, clayton07/qwen2-colpali-ocr, gabrielaltay/vlmqa, gauri-sharan/test-two, huggingface/InferenceSupport/discussions/915, lukiod/dock2, lukiod/streamlit_qwen, lukiod/test2, merve/vision_papers, profchaos/OCR-APP",13
doll-likeness-series_finetunes_20250426_014322.csv_finetunes_20250426_014322.csv ADDED
@@ -0,0 +1,186 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ Kanbara/doll-likeness-series,"---
3
+ license: creativeml-openrail-m
4
+ ---
5
+
6
+
7
+ # Doll Series by Kbr
8
+
9
+ The 'Doll-Series' is a set of LORA focused on realistic Asian faces, with incredible levels of beauty and aesthetics.
10
+
11
+ My Pixiv: https://www.pixiv.net/en/users/92373922
12
+
13
+ My Twitter: https://twitter.com/KbrLoras
14
+
15
+ ---
16
+
17
+ # Table of Contents
18
+
19
+ - [License](#license)
20
+ - [Disclaimer](#disclaimer)
21
+ - [Used Models](#used-models)
22
+ - [LORA Detail](#lora-detail)
23
+ - [KoreanDollLikeness](#koreandolllikeness)
24
+ - [KoreanDollLikeness_v10](#koreandolllikeness_v10)
25
+ - [KoreanDollLikeness_v15](#koreandolllikeness_v15)
26
+ - [KoreanDollLikeness_v20](#koreandolllikeness_v20)
27
+ - [JapaneseDollLikeness](#japanesedolllikeness)
28
+ - [JapaneseDollLikeness_v10](#japanesedolllikeness_v10)
29
+ - [JapaneseDollLikeness_v15](#japanesedolllikeness_v15)
30
+ - [TaiwanDollLikeness](#taiwandolllikeness)
31
+ - [TaiwanDollLikeness_v15](#taiwandolllikeness_v15)
32
+ - [TaiwanDollLikeness_v20](#taiwandolllikeness_v20)
33
+ - [ChinaDollLikeness](#chinadolllikeness)
34
+ - [ChinaDollLikeness_v10](#chinadolllikeness_v10)
35
+ - [ThaiDollLikeness](#thaidolllikeness)
36
+ - [ThaiDollLikeness_v10](#Thaidolllikeness_v10)
37
+
38
+ - [FAQ](#faq)
39
+
40
+
41
+
42
+ ---
43
+ # License
44
+
45
+ This model is open access and available to all, with a CreativeML OpenRAIL-M license further specifying rights and usage.
46
+ The CreativeML OpenRAIL License specifies:
47
+
48
+ 1. You can't use the model to deliberately produce nor share illegal or harmful outputs or content
49
+ 2. The authors claims no rights on the outputs you generate, you are free to use them and are accountable for their use which must not go against the provisions set in the license
50
+ 3. You may re-distribute the weights and use the model commercially and/or as a service. If you do, please be aware you have to include the same use restrictions as the ones in the license and share a copy of the CreativeML OpenRAIL-M to all your users (please read the license entirely and carefully)
51
+ (Full license here: https://huggingface.co/spaces/CompVis/stable-diffusion-license)
52
+
53
+ # Additional Add-ons to license/notes
54
+ 1. You shall take full responsibility for any creative work that uses this model
55
+ 2. Refrain from using this model for malicious intent, harm, defamation, scam or political usages. It may impair and discourage the author from producing more works.
56
+
57
+ # Disclaimer
58
+
59
+ - Creation of SFW and NSFW images is user's decision, user has complete control over to generate NSFW content whether or not.
60
+
61
+ ---
62
+
63
+ # Used Models
64
+
65
+ * Chilled_re_generic_v2
66
+
67
+ - https://github.com/wibus-wee/stable_diffusion_chilloutmix_ipynb
68
+
69
+ * chilloutmix_cilloutmixNi
70
+
71
+ - https://civitai.com/models/6424/chilloutmix
72
+
73
+
74
+ # Recommended settings:
75
+ - Make sure you are aware on the usage instructions of LORA
76
+
77
+ - VAE: vae-ft-mse-840000-ema-pruned (For realistic models)
78
+
79
+ - Sampler: DPM++ SDE Karras (Recommended for best quality, you may try other samplers)
80
+ - Steps: 20 to 35
81
+ - Clipskip: 1 or 2
82
+ - Upscaler : Latent (bicubic antialiased)
83
+ - CFG Scale : 5 to 9
84
+ - LORA weight for txt2img: anywhere between 0.2 to 0.7 are recommended
85
+ - Denoise strength for img2img: 0.4 to 0.7
86
+
87
+
88
+ ---
89
+ # LORA Detail
90
+ ---
91
+ ## KoreanDollLikeness
92
+
93
+ The first version that is widely used by many authors/AI artist/creators
94
+
95
+ ### KoreanDollLikeness_v10
96
+
97
+ - KoreanDollLikeness_v10:
98
+
99
+ <img src=""https://files.catbox.moe/r61ozj.png"" width="""" height="""">
100
+
101
+ ### KoreanDollLikeness_v15
102
+
103
+ - KoreanDollLikeness_v15:
104
+
105
+ <img src=""https://files.catbox.moe/pgcfhc.png"" widht="""" height="""">
106
+
107
+ ### KoreanDollLikeness_v20
108
+
109
+ - KoreanDollLikeness_v20:
110
+
111
+ <img src=""https://files.catbox.moe/thehrt.png"" widht="""" height="""">
112
+
113
+ ---
114
+
115
+ ## JapaneseDollLikeness
116
+
117
+ The Japanese variant version, subjected for v15 in the future
118
+
119
+ ### JapaneseDollLikeness_v10
120
+
121
+ - JapaneseDollLikeness_v10:
122
+
123
+ <img src=""https://files.catbox.moe/cfypot.png"" width="""" height="""">
124
+
125
+ ### JapaneseDollLikeness_v15
126
+
127
+ - JapaneseDollLikeness_v15:
128
+
129
+ <img src=""https://files.catbox.moe/doa0n2.png"" width="""" height="""">
130
+
131
+ ---
132
+
133
+ ## TaiwanDollLikeness
134
+
135
+ The Taiwan variant version, I have decided to discontinue v10, it is still out there somewhere in the internet, you may still find it.
136
+
137
+ ### TaiwanDollLikeness_v15
138
+
139
+ - TaiwanDollLikeness_v15:
140
+
141
+ <img src=""https://files.catbox.moe/5vr2z4.png"" width="""" height="""">
142
+
143
+ ### TaiwanDollLikeness_v20
144
+
145
+ This version is a huge overhaul and remake, instead of building upon v10 or v15, I took small amount of samples from v10 and introduced a new pool of training images.
146
+
147
+ - TaiwanDollLikeness_v20:
148
+
149
+ <img src=""https://files.catbox.moe/f8c9mb.png"" width="""" height="""">
150
+
151
+ ---
152
+
153
+ ## ChinaDollLikeness
154
+
155
+ The China variant version, took awhile despite the requests, will probably make more versions of it in the future.
156
+
157
+ ### ChinaDollLikeness_v10
158
+
159
+ <img src=""https://files.catbox.moe/zpj9ov.png"" width="""" height="""">
160
+
161
+ ---
162
+
163
+ ## ThaiDollLikeness
164
+
165
+ The Thai variant version, took me a long time to make this, many versions were made but this version is the one I've deemed the best out of all. Might update in the future.
166
+
167
+ ### ThaiDollLikeness_v10
168
+
169
+ <img src=""https://files.catbox.moe/imtxsm.png"" width="""" height="""">
170
+
171
+
172
+ # FAQ
173
+ - # Q: Why can't I produce the same pictures as you?
174
+ - A: Sorry I don't share my prompt, you may check the recommended settings, but you may ask me for advice on Twitter or Pixiv
175
+
176
+ - # Q: What is the difference of each version upgrade?
177
+ - A: Version upgrade does not mean it will fix hands or legs, it is mainly difference of the face of the LORA, newer versions have wider range of faces.
178
+
179
+ - # Q: Will you release all your other LORAs?
180
+ - A: Yes, maybe, but I would like to keep certain LORA to be exclusive to fans or supporters in the future through fanbox/ko-fi
181
+
182
+ - # Q: Do you take requests or commissions on making custom LORAs?
183
+ - A: I might do community voting for requests, if you are somehow interested in a custom/exclusive LORA, you may contact my through Pixiv or Twitter for discussion
184
+
185
+ ---
186
+ ","{""id"": ""Kanbara/doll-likeness-series"", ""author"": ""Kanbara"", ""sha"": ""fcbb585f2ce963698354def0166a1ff0980c30b3"", ""last_modified"": ""2023-06-20 12:31:45+00:00"", ""created_at"": ""2023-03-28 13:49:45+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 389, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""license:creativeml-openrail-m"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""license: creativeml-openrail-m"", ""widget_data"": null, ""model_index"": null, ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='ThaiDollLikeness_v10.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='chinaDollLikeness_v10.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='japaneseDollLikeness_v10.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='japaneseDollLikeness_v15.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='koreanDollLikeness_v10.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='koreanDollLikeness_v15.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='koreanDollLikeness_v20.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='taiwanDollLikeness_v15.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='taiwanDollLikeness_v20.safetensors', size=None, blob_id=None, lfs=None)""], ""spaces"": [""pinkqween/DiscordAI""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-06-20 12:31:45+00:00"", ""cardData"": ""license: creativeml-openrail-m"", ""transformersInfo"": null, ""_id"": ""6422f07959a4ff30c277435a"", ""modelId"": ""Kanbara/doll-likeness-series"", ""usedStorage"": 1359994777}",0,,0,,0,,0,,0,"CompVis/stable-diffusion-license, huggingface/InferenceSupport/discussions/new?title=Kanbara/doll-likeness-series&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5BKanbara%2Fdoll-likeness-series%5D(%2FKanbara%2Fdoll-likeness-series)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, pinkqween/DiscordAI",3
elden-ring-diffusion_finetunes_20250426_121513.csv_finetunes_20250426_121513.csv ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ nitrosocke/elden-ring-diffusion,N/A,N/A,0,,0,,0,,0,,0,"CompVis/stable-diffusion-license, Daniela-C/6x_Image_diffusion, Joeythemonster/Text-To-image-AllModels, John6666/Diffusion80XX4sg, John6666/PrintingPress4, John6666/hfd_test_nostopbutton, Nymbo/image_gen_supaqueue, PeepDaSlan9/B2BMGMT_Diffusion60XX, Yntec/PrintingPress, Yntec/ToyWorld, Yntec/blitz_diffusion, huggingface/InferenceSupport/discussions/new?title=nitrosocke/elden-ring-diffusion&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bnitrosocke%2Felden-ring-diffusion%5D(%2Fnitrosocke%2Felden-ring-diffusion)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, phenixrhyder/NSFW-ToyWorld, yangheng/Super-Resolution-Anime-Diffusion",14
falcon-40b-instruct_finetunes_20250424_223250.csv_finetunes_20250424_223250.csv ADDED
@@ -0,0 +1,389 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id,card,metadata,depth,children,children_count,adapters,adapters_count,quantized,quantized_count,merges,merges_count,spaces,spaces_count
2
+ tiiuae/falcon-40b-instruct,"---
3
+ datasets:
4
+ - tiiuae/falcon-refinedweb
5
+ language:
6
+ - en
7
+ inference: false
8
+ license: apache-2.0
9
+ ---
10
+
11
+ # ✨ Falcon-40B-Instruct
12
+
13
+ **Falcon-40B-Instruct is a 40B parameters causal decoder-only model built by [TII](https://www.tii.ae) based on [Falcon-40B](https://huggingface.co/tiiuae/falcon-40b) and finetuned on a mixture of [Baize](https://github.com/project-baize/baize-chatbot). It is made available under the Apache 2.0 license.**
14
+
15
+ *Paper coming soon 😊.*
16
+
17
+ 🤗 To get started with Falcon (inference, finetuning, quantization, etc.), we recommend reading [this great blogpost fron HF](https://huggingface.co/blog/falcon)!
18
+
19
+ ## Why use Falcon-40B-Instruct?
20
+
21
+ * **You are looking for a ready-to-use chat/instruct model based on [Falcon-40B](https://huggingface.co/tiiuae/falcon-40b).**
22
+ * **Falcon-40B is the best open-source model available.** It outperforms [LLaMA](https://github.com/facebookresearch/llama), [StableLM](https://github.com/Stability-AI/StableLM), [RedPajama](https://huggingface.co/togethercomputer/RedPajama-INCITE-Base-7B-v0.1), [MPT](https://huggingface.co/mosaicml/mpt-7b), etc. See the [OpenLLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard).
23
+ * **It features an architecture optimized for inference**, with FlashAttention ([Dao et al., 2022](https://arxiv.org/abs/2205.14135)) and multiquery ([Shazeer et al., 2019](https://arxiv.org/abs/1911.02150)).
24
+
25
+ 💬 **This is an instruct model, which may not be ideal for further finetuning.** If you are interested in building your own instruct/chat model, we recommend starting from [Falcon-40B](https://huggingface.co/tiiuae/falcon-40b).
26
+
27
+ 💸 **Looking for a smaller, less expensive model?** [Falcon-7B-Instruct](https://huggingface.co/tiiuae/falcon-7b-instruct) is Falcon-40B-Instruct's little brother!
28
+
29
+ ```python
30
+ from transformers import AutoTokenizer, AutoModelForCausalLM
31
+ import transformers
32
+ import torch
33
+
34
+ model = ""tiiuae/falcon-40b-instruct""
35
+
36
+ tokenizer = AutoTokenizer.from_pretrained(model)
37
+ pipeline = transformers.pipeline(
38
+ ""text-generation"",
39
+ model=model,
40
+ tokenizer=tokenizer,
41
+ torch_dtype=torch.bfloat16,
42
+ trust_remote_code=True,
43
+ device_map=""auto"",
44
+ )
45
+ sequences = pipeline(
46
+ ""Girafatron is obsessed with giraffes, the most glorious animal on the face of this Earth. Giraftron believes all other animals are irrelevant when compared to the glorious majesty of the giraffe.\nDaniel: Hello, Girafatron!\nGirafatron:"",
47
+ max_length=200,
48
+ do_sample=True,
49
+ top_k=10,
50
+ num_return_sequences=1,
51
+ eos_token_id=tokenizer.eos_token_id,
52
+ )
53
+ for seq in sequences:
54
+ print(f""Result: {seq['generated_text']}"")
55
+
56
+ ```
57
+
58
+ For fast inference with Falcon, check-out [Text Generation Inference](https://github.com/huggingface/text-generation-inference)! Read more in this [blogpost]((https://huggingface.co/blog/falcon).
59
+
60
+ You will need **at least 85-100GB of memory** to swiftly run inference with Falcon-40B.
61
+
62
+
63
+
64
+ # Model Card for Falcon-40B-Instruct
65
+
66
+ ## Model Details
67
+
68
+ ### Model Description
69
+
70
+ - **Developed by:** [https://www.tii.ae](https://www.tii.ae);
71
+ - **Model type:** Causal decoder-only;
72
+ - **Language(s) (NLP):** English and French;
73
+ - **License:** Apache 2.0;
74
+ - **Finetuned from model:** [Falcon-40B](https://huggingface.co/tiiuae/falcon-40b).
75
+
76
+ ### Model Source
77
+
78
+ - **Paper:** *coming soon*.
79
+
80
+ ## Uses
81
+
82
+ ### Direct Use
83
+
84
+ Falcon-40B-Instruct has been finetuned on a chat dataset.
85
+
86
+ ### Out-of-Scope Use
87
+
88
+ Production use without adequate assessment of risks and mitigation; any use cases which may be considered irresponsible or harmful.
89
+
90
+ ## Bias, Risks, and Limitations
91
+
92
+ Falcon-40B-Instruct is mostly trained on English data, and will not generalize appropriately to other languages. Furthermore, as it is trained on a large-scale corpora representative of the web, it will carry the stereotypes and biases commonly encountered online.
93
+
94
+ ### Recommendations
95
+
96
+ We recommend users of Falcon-40B-Instruct to develop guardrails and to take appropriate precautions for any production use.
97
+
98
+ ## How to Get Started with the Model
99
+
100
+
101
+ ```python
102
+ from transformers import AutoTokenizer, AutoModelForCausalLM
103
+ import transformers
104
+ import torch
105
+
106
+ model = ""tiiuae/falcon-40b-instruct""
107
+
108
+ tokenizer = AutoTokenizer.from_pretrained(model)
109
+ pipeline = transformers.pipeline(
110
+ ""text-generation"",
111
+ model=model,
112
+ tokenizer=tokenizer,
113
+ torch_dtype=torch.bfloat16,
114
+ trust_remote_code=True,
115
+ device_map=""auto"",
116
+ )
117
+ sequences = pipeline(
118
+ ""Girafatron is obsessed with giraffes, the most glorious animal on the face of this Earth. Giraftron believes all other animals are irrelevant when compared to the glorious majesty of the giraffe.\nDaniel: Hello, Girafatron!\nGirafatron:"",
119
+ max_length=200,
120
+ do_sample=True,
121
+ top_k=10,
122
+ num_return_sequences=1,
123
+ eos_token_id=tokenizer.eos_token_id,
124
+ )
125
+ for seq in sequences:
126
+ print(f""Result: {seq['generated_text']}"")
127
+
128
+ ```
129
+
130
+ ## Training Details
131
+
132
+ ### Training Data
133
+
134
+ Falcon-40B-Instruct was finetuned on a 150M tokens from [Bai ze](https://github.com/project-baize/baize-chatbot) mixed with 5% of [RefinedWeb](https://huggingface.co/datasets/tiiuae/falcon-refinedweb) data.
135
+
136
+
137
+ The data was tokenized with the Falcon-[7B](https://huggingface.co/tiiuae/falcon-7b)/[40B](https://huggingface.co/tiiuae/falcon-40b) tokenizer.
138
+
139
+
140
+ ## Evaluation
141
+
142
+ *Paper coming soon.*
143
+
144
+ See the [OpenLLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) for early results.
145
+
146
+
147
+ ## Technical Specifications
148
+
149
+ For more information about pretraining, see [Falcon-40B](https://huggingface.co/tiiuae/falcon-40b).
150
+
151
+ ### Model Architecture and Objective
152
+
153
+ Falcon-40B is a causal decoder-only model trained on a causal language modeling task (i.e., predict the next token).
154
+
155
+ The architecture is broadly adapted from the GPT-3 paper ([Brown et al., 2020](https://arxiv.org/abs/2005.14165)), with the following differences:
156
+
157
+ * **Positionnal embeddings:** rotary ([Su et al., 2021](https://arxiv.org/abs/2104.09864));
158
+ * **Attention:** multiquery ([Shazeer et al., 2019](https://arxiv.org/abs/1911.02150)) and FlashAttention ([Dao et al., 2022](https://arxiv.org/abs/2205.14135));
159
+ * **Decoder-block:** parallel attention/MLP with a single layer norm.
160
+
161
+ For multiquery, we are using an internal variant which uses independent key and values per tensor parallel degree.
162
+
163
+ | **Hyperparameter** | **Value** | **Comment** |
164
+ |--------------------|-----------|----------------------------------------|
165
+ | Layers | 60 | |
166
+ | `d_model` | 8192 | |
167
+ | `head_dim` | 64 | Reduced to optimise for FlashAttention |
168
+ | Vocabulary | 65024 | |
169
+ | Sequence length | 2048 | |
170
+
171
+ ### Compute Infrastructure
172
+
173
+ #### Hardware
174
+
175
+ Falcon-40B-Instruct was trained on AWS SageMaker, on 64 A100 40GB GPUs in P4d instances.
176
+
177
+ #### Software
178
+
179
+ Falcon-40B-Instruct was trained a custom distributed training codebase, Gigatron. It uses a 3D parallelism approach combined with ZeRO and high-performance Triton kernels (FlashAttention, etc.)
180
+
181
+
182
+ ## Citation
183
+
184
+ *Paper coming soon* 😊. In the meanwhile, you can use the following information to cite:
185
+ ```
186
+ @article{falcon40b,
187
+ title={{Falcon-40B}: an open large language model with state-of-the-art performance},
188
+ author={Almazrouei, Ebtesam and Alobeidli, Hamza and Alshamsi, Abdulaziz and Cappelli, Alessandro and Cojocaru, Ruxandra and Debbah, Merouane and Goffinet, Etienne and Heslow, Daniel and Launay, Julien and Malartic, Quentin and Noune, Badreddine and Pannier, Baptiste and Penedo, Guilherme},
189
+ year={2023}
190
+ }
191
+ ```
192
+
193
+ To learn more about the pretraining dataset, see the 📓 [RefinedWeb paper](https://arxiv.org/abs/2306.01116).
194
+
195
+ ```
196
+ @article{refinedweb,
197
+ title={The {R}efined{W}eb dataset for {F}alcon {LLM}: outperforming curated corpora with web data, and web data only},
198
+ author={Guilherme Penedo and Quentin Malartic and Daniel Hesslow and Ruxandra Cojocaru and Alessandro Cappelli and Hamza Alobeidli and Baptiste Pannier and Ebtesam Almazrouei and Julien Launay},
199
+ journal={arXiv preprint arXiv:2306.01116},
200
+ eprint={2306.01116},
201
+ eprinttype = {arXiv},
202
+ url={https://arxiv.org/abs/2306.01116},
203
+ year={2023}
204
+ }
205
+ ```
206
+
207
+ To cite the [Baize](https://github.com/project-baize/baize-chatbot) instruction dataset used for this model:
208
+ ```
209
+ @article{xu2023baize,
210
+ title={Baize: An Open-Source Chat Model with Parameter-Efficient Tuning on Self-Chat Data},
211
+ author={Xu, Canwen and Guo, Daya and Duan, Nan and McAuley, Julian},
212
+ journal={arXiv preprint arXiv:2304.01196},
213
+ year={2023}
214
+ }
215
+ ```
216
+
217
+
218
+ ## License
219
+
220
+ Falcon-40B-Instruct is made available under the Apache 2.0 license.
221
+
222
+ ## Contact
223
+ falconllm@tii.ae","{""id"": ""tiiuae/falcon-40b-instruct"", ""author"": ""tiiuae"", ""sha"": ""ecb78d97ac356d098e79f0db222c9ce7c5d9ee5f"", ""last_modified"": ""2023-09-29 14:32:27+00:00"", ""created_at"": ""2023-05-25 10:14:36+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 65164, ""downloads_all_time"": null, ""likes"": 1174, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""pytorch"", ""falcon"", ""text-generation"", ""custom_code"", ""en"", ""dataset:tiiuae/falcon-refinedweb"", ""arxiv:2205.14135"", ""arxiv:1911.02150"", ""arxiv:2005.14165"", ""arxiv:2104.09864"", ""arxiv:2306.01116"", ""arxiv:2304.01196"", ""license:apache-2.0"", ""autotrain_compatible"", ""text-generation-inference"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": ""text-generation"", ""mask_token"": null, ""trending_score"": null, ""card_data"": ""datasets:\n- tiiuae/falcon-refinedweb\nlanguage:\n- en\nlicense: apache-2.0\ninference: false"", ""widget_data"": [{""text"": ""My name is Julien and I like to""}, {""text"": ""I like traveling by train because""}, {""text"": ""Paris is an amazing place to visit,""}, {""text"": ""Once upon a time,""}], ""model_index"": null, ""config"": {""architectures"": [""FalconForCausalLM""], ""auto_map"": {""AutoConfig"": ""configuration_falcon.FalconConfig"", ""AutoModel"": ""modeling_falcon.FalconModel"", ""AutoModelForSequenceClassification"": ""modeling_falcon.FalconForSequenceClassification"", ""AutoModelForTokenClassification"": ""modeling_falcon.FalconForTokenClassification"", ""AutoModelForQuestionAnswering"": ""modeling_falcon.FalconForQuestionAnswering"", ""AutoModelForCausalLM"": ""modeling_falcon.FalconForCausalLM""}, ""model_type"": ""falcon"", ""tokenizer_config"": {""eos_token"": ""<|endoftext|>""}}, ""transformers_info"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='configuration_falcon.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='generation_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='handler.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='modeling_falcon.py', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00001-of-00009.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00002-of-00009.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00003-of-00009.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00004-of-00009.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00005-of-00009.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00006-of-00009.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00007-of-00009.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00008-of-00009.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model-00009-of-00009.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='pytorch_model.bin.index.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)""], ""spaces"": [""Vokturz/can-it-run-llm"", ""HuggingFaceH4/falcon-chat"", ""h2oai/h2ogpt-chatbot"", ""h2oai/h2ogpt-chatbot2"", ""Intel/low_bit_open_llm_leaderboard"", ""BAAI/open_cn_llm_leaderboard"", ""Sharathhebbar24/One-stop-for-Open-source-models"", ""gsaivinay/open_llm_leaderboard"", ""meval/multilingual-chatbot-arena-leaderboard"", ""GTBench/GTBench"", ""Vikhrmodels/small-shlepa-lb"", ""Justinrune/LLaMA-Factory"", ""kenken999/fastapi_django_main_live"", ""kz-transformers/kaz-llm-lb"", ""felixz/open_llm_leaderboard"", ""awacke1/Deepseek-HPC-GPU-KEDA"", ""officialhimanshu595/llama-factory"", ""OPTML-Group/UnlearnCanvas-Benchmark"", ""BAAI/open_flageval_vlm_leaderboard"", ""hlydecker/falcon-chat"", ""b1sheng/kg_llm_leaderboard_test"", ""neubla/neubla-llm-evaluation-board"", ""lambdabrendan/Lambda-LLM-Calculator"", ""lapsapking/h2ogpt-chatbot"", ""bparks08/falcon-chat-40b-1"", ""danfsmithmsft/falcon-chat"", ""NebulaVortex/falcon-chat"", ""rodrigomasini/data_only_open_llm_leaderboard"", ""Docfile/open_llm_leaderboard"", ""his0/h2ogpt-chatbot"", ""atimughal662/InfoFusion"", ""alexkueck/ChatBotLI2Klein"", ""ZENLLC/EagleAsk"", ""HighVibesTimes/falcon-chat"", ""patti-j/Omdena-MHWB-Falcon"", ""qtvhao/falcon-chat"", ""iamrobotbear/falcon-chat"", ""linfso/falcon-chat"", ""bpmf/falcon-chat"", ""Aaksh/tester"", ""ariel0330/h2osiri"", ""bhaskartripathi/PDF-GPT_Falcon"", ""elbanhawy/falcon-chat"", ""Inoob/falcon-chat"", ""Gage888/falcon-Gage-chat-01"", ""ccoreilly/aigua-xat"", ""Felix9390/Graphica-2"", ""Sambhavnoobcoder/h2ogpt-chatbot"", ""adnansami1992sami/falcon-chat"", ""dzekoh/ChatBot"", ""k3ybladewielder/gen_app_chat"", ""E1829/Falcon"", ""E1829/falcon-chat2"", ""busraasan/gpt_comparison"", ""k3ybladewielder/falcon-40b-instruct"", ""smothiki/open_llm_leaderboard"", ""mouliraj56/falcon"", ""blackwingedkite/gutalk"", ""mouliraj56/testpdf"", ""pngwn/open_llm_leaderboard"", ""pngwn/open_llm_leaderboard_two"", ""yourstudiolab/eai1a"", ""iblfe/test"", ""watsonswx/chatrobot"", ""UdayG01/mockinterview-v1"", ""aimaswx/chatbot_v1"", ""AnonymousSub/Ayurveda_Chatbot"", ""K00B404/Research-chatbot"", ""iShare/pdf_ai_bot_hf"", ""Cran-May/falcon-40b-instruct-ggml"", ""blackwingedkite/alpaca2_clas"", ""wissamantoun/LLM_Detection_Attribution"", ""0x1668/open_llm_leaderboard"", ""pngwn/open_llm_leaderboard-check"", ""asir0z/open_llm_leaderboard"", ""kelvin-t-lu/chatbot"", ""Nymbo/can-it-run-llm"", ""Elreniel/llm_chat"", ""kbmlcoding/open_llm_leaderboard_free"", ""acecalisto3/SAIOPSYS"", ""acloudfan/HF-Playground"", ""binqiangliu/DocChat_WM"", ""Rehman1603/Chat_with_Document"", ""dbasu/multilingual-chatbot-arena-leaderboard"", ""PrathamSharma/dima806-indian_food_image_detection"", ""K00B404/One-stop-till-you-drop"", ""aichampions/open_llm_leaderboard"", ""Adeco/open_llm_leaderboard"", ""RobinsAIWorld/can-it-run-llm"", ""Jamiiwej2903/kaaaa"", ""anirudh937/open_llm_leaderboard"", ""smothiki/open_llm_leaderboard2"", ""cw332/h2ogpt-chatbot"", ""samkn123/chatbot"", ""Asim07/cover-query"", ""mjalg/IFEvalTR"", ""mani1996123/newkjnkjn"", ""Shrijoy/falcon_summarizer"", ""abugaber/test"", ""mpvasilis/can-it-run-llm""], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-09-29 14:32:27+00:00"", ""cardData"": ""datasets:\n- tiiuae/falcon-refinedweb\nlanguage:\n- en\nlicense: apache-2.0\ninference: false"", ""transformersInfo"": {""auto_model"": ""AutoModelForCausalLM"", ""custom_class"": null, ""pipeline_tag"": ""text-generation"", ""processor"": ""AutoTokenizer""}, ""_id"": ""646f350ce2a72c647b686b91"", ""modelId"": ""tiiuae/falcon-40b-instruct"", ""usedStorage"": 167352935898}",0,"https://huggingface.co/lucas0/empath-falcon-40b, https://huggingface.co/anirudh2403/results, https://huggingface.co/miragexy/ins_fine_tuned_falcon40B",3,"https://huggingface.co/mubashirsaeed/care-bot-harry-potter-falcon-40b, https://huggingface.co/Shishir17/falcon40binstruct_mentalhealthmodel_oct23",2,"https://huggingface.co/mradermacher/falcon-40b-instruct-GGUF, https://huggingface.co/mradermacher/falcon-40b-instruct-i1-GGUF",2,,0,"BAAI/open_cn_llm_leaderboard, GTBench/GTBench, HuggingFaceH4/falcon-chat, HuggingFaceH4/open_llm_leaderboard, Intel/low_bit_open_llm_leaderboard, Justinrune/LLaMA-Factory, OPTML-Group/UnlearnCanvas-Benchmark, Sharathhebbar24/One-stop-for-Open-source-models, Vikhrmodels/small-shlepa-lb, Vokturz/can-it-run-llm, awacke1/Deepseek-HPC-GPU-KEDA, huggingface/InferenceSupport/discussions/new?title=tiiuae/falcon-40b-instruct&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Btiiuae%2Ffalcon-40b-instruct%5D(%2Ftiiuae%2Ffalcon-40b-instruct)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A, kenken999/fastapi_django_main_live, meval/multilingual-chatbot-arena-leaderboard",14
224
+ lucas0/empath-falcon-40b,"---
225
+ license: apache-2.0
226
+ base_model: tiiuae/falcon-40b-instruct
227
+ tags:
228
+ - generated_from_trainer
229
+ model-index:
230
+ - name: empath-falcon-40b
231
+ results: []
232
+ ---
233
+
234
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
235
+ should probably proofread and complete it, then remove this comment. -->
236
+
237
+ # empath-falcon-40b
238
+
239
+ This model is a fine-tuned version of [tiiuae/falcon-40b-instruct](https://huggingface.co/tiiuae/falcon-40b-instruct) on the None dataset.
240
+
241
+ ## Model description
242
+
243
+ More information needed
244
+
245
+ ## Intended uses & limitations
246
+
247
+ More information needed
248
+
249
+ ## Training and evaluation data
250
+
251
+ More information needed
252
+
253
+ ## Training procedure
254
+
255
+ ### Training hyperparameters
256
+
257
+ The following hyperparameters were used during training:
258
+ - learning_rate: 0.0002
259
+ - train_batch_size: 4
260
+ - eval_batch_size: 8
261
+ - seed: 42
262
+ - gradient_accumulation_steps: 4
263
+ - total_train_batch_size: 16
264
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
265
+ - lr_scheduler_type: cosine
266
+ - lr_scheduler_warmup_ratio: 0.05
267
+ - num_epochs: 3
268
+
269
+ ### Training results
270
+
271
+
272
+
273
+ ### Framework versions
274
+
275
+ - Transformers 4.32.0.dev0
276
+ - Pytorch 2.0.1+cu117
277
+ - Datasets 2.13.1
278
+ - Tokenizers 0.13.3
279
+ ","{""id"": ""lucas0/empath-falcon-40b"", ""author"": ""lucas0"", ""sha"": ""826c4a41158acd1c5846aff0d104ac116885b36f"", ""last_modified"": ""2023-07-28 01:04:53+00:00"", ""created_at"": ""2023-07-20 18:29:06+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": null, ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""generated_from_trainer"", ""base_model:tiiuae/falcon-40b-instruct"", ""base_model:finetune:tiiuae/falcon-40b-instruct"", ""license:apache-2.0"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: tiiuae/falcon-40b-instruct\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: empath-falcon-40b\n results: []"", ""widget_data"": null, ""model_index"": [{""name"": ""empath-falcon-40b"", ""results"": []}], ""config"": null, ""transformers_info"": null, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='.gitignore', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='training_args.bin', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-07-28 01:04:53+00:00"", ""cardData"": ""base_model: tiiuae/falcon-40b-instruct\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: empath-falcon-40b\n results: []"", ""transformersInfo"": null, ""_id"": ""64b97cf284ddd52599b8cde3"", ""modelId"": ""lucas0/empath-falcon-40b"", ""usedStorage"": 133840248}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=lucas0/empath-falcon-40b&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Blucas0%2Fempath-falcon-40b%5D(%2Flucas0%2Fempath-falcon-40b)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
280
+ anirudh2403/results,"---
281
+ license: apache-2.0
282
+ base_model: tiiuae/falcon-40b-instruct
283
+ tags:
284
+ - generated_from_trainer
285
+ model-index:
286
+ - name: results
287
+ results: []
288
+ ---
289
+
290
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
291
+ should probably proofread and complete it, then remove this comment. -->
292
+
293
+ # results
294
+
295
+ This model is a fine-tuned version of [tiiuae/falcon-40b-instruct](https://huggingface.co/tiiuae/falcon-40b-instruct) on an unknown dataset.
296
+
297
+ ## Model description
298
+
299
+ More information needed
300
+
301
+ ## Intended uses & limitations
302
+
303
+ More information needed
304
+
305
+ ## Training and evaluation data
306
+
307
+ More information needed
308
+
309
+ ## Training procedure
310
+
311
+ ### Training hyperparameters
312
+
313
+ The following hyperparameters were used during training:
314
+ - learning_rate: 0.0002
315
+ - train_batch_size: 4
316
+ - eval_batch_size: 8
317
+ - seed: 42
318
+ - gradient_accumulation_steps: 4
319
+ - total_train_batch_size: 16
320
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
321
+ - lr_scheduler_type: constant
322
+ - lr_scheduler_warmup_ratio: 0.03
323
+ - training_steps: 500
324
+
325
+ ### Framework versions
326
+
327
+ - Transformers 4.33.2
328
+ - Pytorch 2.0.1+cu118
329
+ - Datasets 2.14.5
330
+ - Tokenizers 0.13.3
331
+ ","{""id"": ""anirudh2403/results"", ""author"": ""anirudh2403"", ""sha"": ""1f056d7d68f8d4f42e40fa10ae66483ffc2c5662"", ""last_modified"": ""2023-09-19 20:17:00+00:00"", ""created_at"": ""2023-09-19 19:09:48+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 6, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""generated_from_trainer"", ""base_model:tiiuae/falcon-40b-instruct"", ""base_model:finetune:tiiuae/falcon-40b-instruct"", ""license:apache-2.0"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: tiiuae/falcon-40b-instruct\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: results\n results: []"", ""widget_data"": null, ""model_index"": [{""name"": ""results"", ""results"": []}], ""config"": {""tokenizer_config"": {""eos_token"": ""<|endoftext|>""}}, ""transformers_info"": {""auto_model"": ""AutoModel"", ""custom_class"": null, ""pipeline_tag"": null, ""processor"": null}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_model.bin', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='training_args.bin', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2023-09-19 20:17:00+00:00"", ""cardData"": ""base_model: tiiuae/falcon-40b-instruct\nlicense: apache-2.0\ntags:\n- generated_from_trainer\nmodel-index:\n- name: results\n results: []"", ""transformersInfo"": {""auto_model"": ""AutoModel"", ""custom_class"": null, ""pipeline_tag"": null, ""processor"": null}, ""_id"": ""6509f1fccc02352e1e28b26d"", ""modelId"": ""anirudh2403/results"", ""usedStorage"": 1777517192}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=anirudh2403/results&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Banirudh2403%2Fresults%5D(%2Fanirudh2403%2Fresults)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
332
+ miragexy/ins_fine_tuned_falcon40B,"---
333
+ base_model: tiiuae/falcon-40b-instruct
334
+ library_name: transformers
335
+ model_name: ins_fine_tuned_falcon40B
336
+ tags:
337
+ - generated_from_trainer
338
+ - trl
339
+ - sft
340
+ licence: license
341
+ ---
342
+
343
+ # Model Card for ins_fine_tuned_falcon40B
344
+
345
+ This model is a fine-tuned version of [tiiuae/falcon-40b-instruct](https://huggingface.co/tiiuae/falcon-40b-instruct).
346
+ It has been trained using [TRL](https://github.com/huggingface/trl).
347
+
348
+ ## Quick start
349
+
350
+ ```python
351
+ from transformers import pipeline
352
+
353
+ question = ""If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?""
354
+ generator = pipeline(""text-generation"", model=""miragexy/ins_fine_tuned_falcon40B"", device=""cuda"")
355
+ output = generator([{""role"": ""user"", ""content"": question}], max_new_tokens=128, return_full_text=False)[0]
356
+ print(output[""generated_text""])
357
+ ```
358
+
359
+ ## Training procedure
360
+
361
+ [<img src=""https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg"" alt=""Visualize in Weights & Biases"" width=""150"" height=""24""/>](https://wandb.ai/a7med7usam777-technology-innovation-institute/huggingface/runs/m0vfp2ng)
362
+
363
+
364
+ This model was trained with SFT.
365
+
366
+ ### Framework versions
367
+
368
+ - TRL: 0.13.0
369
+ - Transformers: 4.48.0
370
+ - Pytorch: 2.5.1
371
+ - Datasets: 3.2.0
372
+ - Tokenizers: 0.21.0
373
+
374
+ ## Citations
375
+
376
+
377
+
378
+ Cite TRL as:
379
+
380
+ ```bibtex
381
+ @misc{vonwerra2022trl,
382
+ title = {{TRL: Transformer Reinforcement Learning}},
383
+ author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
384
+ year = 2020,
385
+ journal = {GitHub repository},
386
+ publisher = {GitHub},
387
+ howpublished = {\url{https://github.com/huggingface/trl}}
388
+ }
389
+ ```","{""id"": ""miragexy/ins_fine_tuned_falcon40B"", ""author"": ""miragexy"", ""sha"": ""746d28f543a4c824a11e4731428610eeef07e4fb"", ""last_modified"": ""2025-01-12 15:18:09+00:00"", ""created_at"": ""2025-01-09 20:46:14+00:00"", ""private"": false, ""gated"": false, ""disabled"": false, ""downloads"": 0, ""downloads_all_time"": null, ""likes"": 0, ""library_name"": ""transformers"", ""gguf"": null, ""inference"": null, ""inference_provider_mapping"": null, ""tags"": [""transformers"", ""tensorboard"", ""safetensors"", ""generated_from_trainer"", ""trl"", ""sft"", ""base_model:tiiuae/falcon-40b-instruct"", ""base_model:finetune:tiiuae/falcon-40b-instruct"", ""endpoints_compatible"", ""region:us""], ""pipeline_tag"": null, ""mask_token"": null, ""trending_score"": null, ""card_data"": ""base_model: tiiuae/falcon-40b-instruct\nlibrary_name: transformers\nmodel_name: ins_fine_tuned_falcon40B\ntags:\n- generated_from_trainer\n- trl\n- sft\nlicence: license"", ""widget_data"": null, ""model_index"": null, ""config"": {""tokenizer_config"": {""eos_token"": ""<|endoftext|>"", ""pad_token"": ""<|endoftext|>""}}, ""transformers_info"": {""auto_model"": ""AutoModel"", ""custom_class"": null, ""pipeline_tag"": null, ""processor"": null}, ""siblings"": [""RepoSibling(rfilename='.gitattributes', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='README.md', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='adapter_model.safetensors', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='logs/events.out.tfevents.1736455575.c575303a0d41.1684.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='logs/events.out.tfevents.1736456477.c575303a0d41.9549.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='logs/events.out.tfevents.1736458064.c575303a0d41.16633.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='logs/events.out.tfevents.1736634088.192-222-59-245.6243.0', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='special_tokens_map.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='tokenizer_config.json', size=None, blob_id=None, lfs=None)"", ""RepoSibling(rfilename='training_args.bin', size=None, blob_id=None, lfs=None)""], ""spaces"": [], ""safetensors"": null, ""security_repo_status"": null, ""xet_enabled"": null, ""lastModified"": ""2025-01-12 15:18:09+00:00"", ""cardData"": ""base_model: tiiuae/falcon-40b-instruct\nlibrary_name: transformers\nmodel_name: ins_fine_tuned_falcon40B\ntags:\n- generated_from_trainer\n- trl\n- sft\nlicence: license"", ""transformersInfo"": {""auto_model"": ""AutoModel"", ""custom_class"": null, ""pipeline_tag"": null, ""processor"": null}, ""_id"": ""67803596a1ab15dbf8250d1b"", ""modelId"": ""miragexy/ins_fine_tuned_falcon40B"", ""usedStorage"": 267498174}",1,,0,,0,,0,,0,huggingface/InferenceSupport/discussions/new?title=miragexy/ins_fine_tuned_falcon40B&description=React%20to%20this%20comment%20with%20an%20emoji%20to%20vote%20for%20%5Bmiragexy%2Fins_fine_tuned_falcon40B%5D(%2Fmiragexy%2Fins_fine_tuned_falcon40B)%20to%20be%20supported%20by%20Inference%20Providers.%0A%0A(optional)%20Which%20providers%20are%20you%20interested%20in%3F%20(Novita%2C%20Hyperbolic%2C%20Together%E2%80%A6)%0A,1
flan-t5-large_finetunes_20250424_223250.csv_finetunes_20250424_223250.csv ADDED
The diff for this file is too large to render. See raw diff