Upload folder using huggingface_hub

README.md

@@ -1,7 +1,6 @@
 ---
 license: apache-2.0
-base_model:
-- lightonai/LightOnOCR-2-1B
+pipeline_tag: image-text-to-text
 language:
 - en
 - fr
@@ -12,8 +11,8 @@ language:
 - pt
 - sv
 - da
-- ja
 - zh
+- ja
 library_name: transformers
 tags:
 - ocr
@@ -23,7 +22,6 @@ tags:
 - tables
 - forms
 - mlx
-pipeline_tag: image-text-to-text
 ---
 
 # mlx-community/LightOnOCR-2-1B-4bit
@@ -31,17 +29,10 @@ This model was converted to MLX format from [`lightonai/LightOnOCR-2-1B`]() usin
 Refer to the [original model card](https://huggingface.co/lightonai/LightOnOCR-2-1B) for more details on the model.
 ## Use with mlx
 
-> **Note:** LightOnOCR-2 requires transformers installed from source (not yet in a stable release).
-
-```bash
-pip install git+https://github.com/huggingface/transformers
-pip install pillow pypdfium2
-```
-
 ```bash
 pip install -U mlx-vlm
 ```
 
 ```bash
-python -m mlx_vlm.generate --model mlx-community/LightOnOCR-2-1B-4bit --max-tokens 100 --temperature 0.0 --prompt "" --image <path_to_image>
-```
+python -m mlx_vlm.generate --model mlx-community/LightOnOCR-2-1B-4bit --max-tokens 100 --temperature 0.0 --prompt "Describe this image." --image <path_to_image>
+```

processor_config.json

@@ -2,15 +2,10 @@
   "image_break_token": "<|vision_pad|>",
   "image_end_token": "<|vision_end|>",
   "image_processor": {
-    "crop_size": null,
     "data_format": "channels_first",
     "default_to_square": true,
-    "device": null,
-    "disable_grouping": null,
-    "do_center_crop": null,
     "do_convert_rgb": true,
     "do_normalize": true,
-    "do_pad": null,
     "do_rescale": true,
     "do_resize": true,
     "image_mean": [
@@ -19,19 +14,14 @@
       0.40821073
     ],
     "image_processor_type": "PixtralImageProcessorFast",
-    "image_seq_length": null,
     "image_std": [
       0.26862954,
       0.26130258,
       0.27577711
     ],
-    "input_data_format": null,
-    "pad_size": null,
     "patch_size": 14,
-    "processor_class": "PixtralProcessor",
     "resample": 3,
     "rescale_factor": 0.00392156862745098,
-    "return_tensors": null,
     "size": {
       "longest_edge": 1540
     }

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:26628a6864bf44c17c7f4c42a68b37a43574a94420d094931580bbb0d5e8b84e
-size 11422632
+oid sha256:f54b55fa0c3aba0c91ce09ea79ae4e62da24e2a1a630f96c4bae34aba25e234a
+size 11422822

tokenizer_config.json

@@ -1,20 +1,5 @@
 {
   "add_prefix_space": false,
-  "additional_special_tokens": [
-    "<|im_start|>",
-    "<|im_end|>",
-    "<|object_ref_start|>",
-    "<|object_ref_end|>",
-    "<|box_start|>",
-    "<|box_end|>",
-    "<|quad_start|>",
-    "<|quad_end|>",
-    "<|vision_start|>",
-    "<|vision_end|>",
-    "<|vision_pad|>",
-    "<|image_pad|>",
-    "<|video_pad|>"
-  ],
   "backend": "tokenizers",
   "bos_token": null,
   "clean_up_tokenization_spaces": false,
@@ -39,12 +24,15 @@
   "image_end_token": "<|vision_end|>",
   "image_token": "<|image_pad|>",
   "is_local": true,
+  "max_length": null,
   "model_max_length": 131072,
   "model_specific_special_tokens": {},
+  "pad_to_multiple_of": null,
   "pad_token": "<|endoftext|>",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
   "processor_class": "PixtralProcessor",
   "split_special_tokens": false,
-  "tokenizer_class": "LlamaTokenizer",
-  "unk_token": null,
-  "use_default_system_prompt": false
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": null
 }