Hugging Face's logo

EZCon
/
GLM-OCR-4bit-mlx

Image-Text-to-Text
MLX
Safetensors
glm_ocr
conversational
4-bit precision
Model card Files
xet
 Community
hermeschen-ezcon commited on
Commit
4e8db63
·
verified ·
1 Parent(s): 10b1b7a

Upload folder using huggingface_hub

Browse files
Files changed (2) hide show
  1. README.md +6 -4
  2. processor_config.json +1 -2
README.md CHANGED
@@ -10,16 +10,18 @@ language:
10
  - ja
11
  - ko
12
  pipeline_tag: image-to-text
13
- library_name: transformers
14
- base_model:
15
- - zai-org/GLM-OCR
16
  tags:
17
  - mlx
18
  ---
19
 
20
  # EZCon/GLM-OCR-4bit-mlx
21
- This model was converted to MLX format from [`unsloth/GLM-OCR`]() using mlx-vlm version **0.4.1**.
 
 
22
  Refer to the [original model card](https://huggingface.co/unsloth/GLM-OCR) for more details on the model.
 
23
  ## Use with mlx
24
 
25
  ```bash
 
10
  - ja
11
  - ko
12
  pipeline_tag: image-to-text
13
+ library_name: mlx
14
+ base_model: unsloth/GLM-OCR
 
15
  tags:
16
  - mlx
17
  ---
18
 
19
  # EZCon/GLM-OCR-4bit-mlx
20
+
21
+ This model was converted to MLX format from [`unsloth/GLM-OCR`](https://huggingface.co/unsloth/GLM-OCR)
22
+ using mlx-vlm version **0.4.2**.
23
  Refer to the [original model card](https://huggingface.co/unsloth/GLM-OCR) for more details on the model.
24
+
25
  ## Use with mlx
26
 
27
  ```bash
processor_config.json CHANGED
@@ -1,6 +1,5 @@
1
  {
2
  "image_processor": {
3
- "data_format": "channels_first",
4
  "do_convert_rgb": true,
5
  "do_normalize": true,
6
  "do_rescale": true,
@@ -10,7 +9,7 @@
10
  0.4578275,
11
  0.40821073
12
  ],
13
- "image_processor_type": "Glm46VImageProcessorFast",
14
  "image_std": [
15
  0.26862954,
16
  0.26130258,
 
1
  {
2
  "image_processor": {
 
3
  "do_convert_rgb": true,
4
  "do_normalize": true,
5
  "do_rescale": true,
 
9
  0.4578275,
10
  0.40821073
11
  ],
12
+ "image_processor_type": "Glm46VImageProcessor",
13
  "image_std": [
14
  0.26862954,
15
  0.26130258,