---
license: apache-2.0
base_model: allenai/olmOCR-2-7B-1025
tags:
- mlx
- vision
- ocr
- quantized
- apple-silicon
---

# olmOCR-2-7B-1025-MLX-8bit

This is an 8-bit quantized version of [allenai/olmOCR-2-7B-1025](https://huggingface.co/allenai/olmOCR-2-7B-1025) optimized for Apple Silicon using MLX.

## Model Description

olmOCR-2 is a state-of-the-art OCR (Optical Character Recognition) vision-language model fine-tuned from Qwen2.5-VL-7B-Instruct. This 8-bit quantized version provides excellent quality with significantly reduced memory footprint.

**Base Model:** allenai/olmOCR-2-7B-1025  
**Quantization:** 8-bit using MLX  
**Model Size:** 8.4 GB (down from ~14 GB BF16)  
**Size Reduction:** ~40%  

## Performance

olmOCR-2 achieves **82.4 points on olmOCR-Bench**, representing state-of-the-art performance for real-world OCR of English-language digitized print documents. The model has been additionally fine-tuned using GRPO RL training to boost performance on:
- Math equations
- Tables
- Complex layouts
- Handwriting

## Usage

### Requirements

```bash
pip install mlx-vlm
```

### Basic Usage

```python
from mlx_vlm import load, generate
from PIL import Image

# Load the model
model, processor = load("richardyoung/olmOCR-2-7B-1025-MLX-8bit")

# Load your image
image = Image.open("document.png")

# Extract text
prompt = "Extract all text from this image."
output = generate(model, processor, image, prompt, max_tokens=2048)
print(output)
```

### Command Line

```bash
python -m mlx_vlm.generate \
  --model richardyoung/olmOCR-2-7B-1025-MLX-8bit \
  --image document.png \
  --prompt "Extract all text from this image." \
  --max-tokens 2048
```

## Quantization Details

- **Method:** MLX native quantization
- **Bits:** 8-bit
- **Group Size:** Default
- **Recommended for:** Users who prioritize quality and have sufficient RAM (10GB+)

## Model Variants

| Variant | Size | Precision | Use Case |
|---------|------|-----------|----------|
| [8-bit](https://huggingface.co/richardyoung/olmOCR-2-7B-1025-MLX-8bit) | 8.4 GB | Highest | Best quality, more RAM |
| [6-bit](https://huggingface.co/richardyoung/olmOCR-2-7B-1025-MLX-6bit) | 6.4 GB | High | Balanced quality/size |
| [4-bit](https://huggingface.co/richardyoung/olmOCR-2-7B-1025-MLX-4bit) | 4.5 GB | Good | Smallest size, less RAM |

## System Requirements

- **Platform:** Apple Silicon (M1/M2/M3/M4)
- **RAM:** 10+ GB recommended
- **OS:** macOS 12.0+

## Limitations

- Optimized primarily for English-language printed documents
- May have reduced performance on handwritten text compared to printed text
- Requires Apple Silicon hardware for optimal performance

## Citation

```bibtex
@article{olmocr2,
  title={olmOCR 2: Unit test rewards for document OCR},
  author={Allen Institute for AI},
  year={2025}
}
```

## License

Apache 2.0 (inherited from base model)

## Acknowledgements

- Base model by [Allen Institute for AI](https://allenai.org/)
- Quantized for MLX by richardyoung
- Built with [MLX-VLM](https://github.com/Blaizzy/mlx-vlm)

---

*Generated with Claude Code*