Hugging Face's logo

nvlan
/
ndc

Transformers
Safetensors
English
text-generation-inference
unsloth
qwen3_vl
Model card Files
xet
 Community
nvlan commited on
Commit
8e2a9c1
·
verified ·
1 Parent(s): 7169ef0

Upload model trained with Unsloth

Browse files

Upload model trained with Unsloth 2x faster

Files changed (5) hide show
  1. config.json +81 -0
  2. generation_config.json +13 -0
  3. model-00001-of-00002.safetensors +3 -0
  4. model-00002-of-00002.safetensors +3 -0
  5. model.safetensors.index.json +0 -0
config.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3VLForConditionalGeneration"
4
+ ],
5
+ "dtype": "float16",
6
+ "eos_token_id": 151645,
7
+ "image_token_id": 151655,
8
+ "model_name": "unsloth/Qwen3-VL-8B-Instruct-unsloth-bnb-4bit",
9
+ "model_type": "qwen3_vl",
10
+ "pad_token_id": 151654,
11
+ "quantization_config": {
12
+ "bnb_4bit_compute_dtype": "float16",
13
+ "bnb_4bit_quant_type": "nf4",
14
+ "bnb_4bit_use_double_quant": true,
15
+ "llm_int8_enable_fp32_cpu_offload": false,
16
+ "llm_int8_has_fp16_weight": false,
17
+ "llm_int8_skip_modules": null,
18
+ "llm_int8_threshold": 6.0,
19
+ "load_in_4bit": true,
20
+ "load_in_8bit": false,
21
+ "quant_method": "bitsandbytes"
22
+ },
23
+ "text_config": {
24
+ "attention_bias": false,
25
+ "attention_dropout": 0.0,
26
+ "bos_token_id": 151643,
27
+ "dtype": "float16",
28
+ "eos_token_id": 151645,
29
+ "head_dim": 128,
30
+ "hidden_act": "silu",
31
+ "hidden_size": 4096,
32
+ "initializer_range": 0.02,
33
+ "intermediate_size": 12288,
34
+ "max_position_embeddings": 262144,
35
+ "model_type": "qwen3_vl_text",
36
+ "num_attention_heads": 32,
37
+ "num_hidden_layers": 36,
38
+ "num_key_value_heads": 8,
39
+ "rms_norm_eps": 1e-06,
40
+ "rope_scaling": {
41
+ "mrope_interleaved": true,
42
+ "mrope_section": [
43
+ 24,
44
+ 20,
45
+ 20
46
+ ],
47
+ "rope_type": "default"
48
+ },
49
+ "rope_theta": 5000000,
50
+ "use_cache": true,
51
+ "vocab_size": 151936
52
+ },
53
+ "tie_word_embeddings": false,
54
+ "transformers_version": "4.57.1",
55
+ "unsloth_fixed": true,
56
+ "unsloth_version": "2026.3.3",
57
+ "video_token_id": 151656,
58
+ "vision_config": {
59
+ "deepstack_visual_indexes": [
60
+ 8,
61
+ 16,
62
+ 24
63
+ ],
64
+ "depth": 27,
65
+ "dtype": "float16",
66
+ "hidden_act": "gelu_pytorch_tanh",
67
+ "hidden_size": 1152,
68
+ "in_channels": 3,
69
+ "initializer_range": 0.02,
70
+ "intermediate_size": 4304,
71
+ "model_type": "qwen3_vl",
72
+ "num_heads": 16,
73
+ "num_position_embeddings": 2304,
74
+ "out_hidden_size": 4096,
75
+ "patch_size": 16,
76
+ "spatial_merge_size": 2,
77
+ "temporal_patch_size": 2
78
+ },
79
+ "vision_end_token_id": 151653,
80
+ "vision_start_token_id": 151652
81
+ }
generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 151645,
6
+ 151643
7
+ ],
8
+ "pad_token_id": 151654,
9
+ "temperature": 0.7,
10
+ "top_k": 20,
11
+ "top_p": 0.8,
12
+ "transformers_version": "4.57.1"
13
+ }
model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fbe84925aceb9ad8f2e747615ea9514e7d8f718867dda74f6d836d8d85ddc4a9
3
+ size 4984017113
model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:239259c2c5808c241870a6e0b31355c9ab926e3a21d7b7c0dc8f528b6199231d
3
+ size 2715459852
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff