---
base_model: baidu/Unlimited-OCR
license: mit
language:
- multilingual
pipeline_tag: image-text-to-text
tags:
- mlx
- mlx-vlm
- ocr
- vision-language
- baidu
library_name: mlx
---

# Unlimited-OCR BF16 MLX

MLX conversion of [`baidu/Unlimited-OCR`](https://huggingface.co/baidu/Unlimited-OCR).

```python
from mlx_vlm import load
from mlx_vlm.generate import generate

model, processor = load("mikoy92/Unlimited-OCR-bf16-mlx")
result = generate(
    model,
    processor,
    "<image>\ndocument parsing.",
    image="your_image.jpg",
    max_tokens=512,
    temperature=0.0,
)
print(result.text)
```

CLI:

```bash
python -m mlx_vlm generate \
  --model mikoy92/Unlimited-OCR-bf16-mlx \
  --image your_image.jpg \
  --prompt "document parsing." \
  --temp 0
```

This repo stores MLX-layout weights in safetensors (`format=mlx`). It uses the existing `deepseekocr` MLX implementation because Unlimited-OCR shares that SAM + CLIP-L + DeepSeekV2 OCR architecture shape with different checkpoint dimensions.