Churro

🤗 Model • 🗂️ Dataset • 📄 Paper

📚 Docs • 🏆 Leaderboard •

Churro is the fastest way to turn hard-to-read historical scans into reliable text. It gives researchers, libraries, archives, and product teams a unified OCR toolkit for handwritten and printed sources, combining high accuracy, low operating cost, and a clean Python API and CLI workflow.

Supported OCR Models and Backends

We provide first-party support for Churro VLM, the best OCR model for historical documents.

Churro also includes built-in profiles, templates, and post-processing for many other models and integrations, including:

• Hosted vision-language models, including Gemini, GPT, Claude, and more, through LiteLLM integration
• OpenAI-compatible servers, including vLLM, Ollama, TGI, and more
• Azure Document Intelligence
• Mistral OCR
• Chandra OCR
• DeepSeek OCR
• Dots OCR
• MinerU
• Infinity Parser
• PaddleOCR VL
• LFM VL

Quick Start

Python 3.12+ and uv are required.

uv tool install churro-ocr
churro-ocr install hf
churro-ocr transcribe --image scan.png --backend hf --model stanford-oval/churro-3B

For more in-depth information, see the Getting Started guide.

Churro Model and Dataset

• Churro 3B VLM exceeds the accuracy of Gemini 2.5 Pro at 15.5x lower cost.
• Churro-DS dataset contains ~100K pages from 155 historical collections spanning 22 centuries and 46 language clusters.

_{Cost vs. accuracy: Churro (3B) achieves higher accuracy than much larger commercial and open-weight VLMs while being substantially cheaper.}

The following are pages from the CHURRO dev set, randomly picked from the subset where Churro outperforms Gemini 2.5 Pro on the main metric, Normalized Levenshtein Similarity (NLS).

Arabic handwriting Churro 93.9 vs Gemini 92.3 NLS	Bangla print Churro 91.4 vs Gemini 84.3 NLS	Bulgarian print Churro 99.8 vs Gemini 99.2 NLS	Catalan handwriting Churro 95.2 vs Gemini 94.1 NLS	Chinese handwriting Churro 100.0 vs Gemini 95.0 NLS
Czech print Churro 95.9 vs Gemini 95.3 NLS	Dutch print Churro 98.7 vs Gemini 98.0 NLS	English handwriting Churro 99.8 vs Gemini 99.4 NLS	Finnish print Churro 99.6 vs Gemini 99.5 NLS	French handwriting Churro 92.6 vs Gemini 90.6 NLS
German handwriting Churro 81.4 vs Gemini 45.6 NLS	Greek handwriting Churro 81.2 vs Gemini 71.2 NLS	Hebrew handwriting Churro 90.0 vs Gemini 21.6 NLS	Hindi print Churro 98.1 vs Gemini 88.0 NLS	Italian handwriting Churro 93.3 vs Gemini 88.3 NLS
Japanese handwriting Churro 68.9 vs Gemini 13.8 NLS	Khmer handwriting Churro 27.7 vs Gemini 23.3 NLS	Latin handwriting Churro 75.1 vs Gemini 58.3 NLS	Norwegian handwriting Churro 69.7 vs Gemini 65.3 NLS	Persian handwriting Churro 77.6 vs Gemini 74.6 NLS
Polish print Churro 84.4 vs Gemini 0.0 NLS	Portuguese handwriting Churro 52.0 vs Gemini 51.6 NLS	Romanian print Churro 90.9 vs Gemini 45.7 NLS	Sanskrit print Churro 97.5 vs Gemini 97.0 NLS	Slovenian print Churro 98.7 vs Gemini 98.5 NLS
Spanish print Churro 97.9 vs Gemini 78.5 NLS	Swedish handwriting Churro 87.1 vs Gemini 85.1 NLS	Turkish handwriting Churro 74.1 vs Gemini 42.9 NLS	Vietnamese handwriting Churro 87.6 vs Gemini 86.0 NLS

Citation

If you use CHURRO or CHURRO-DS, please cite:

@inproceedings{semnani2025churro,
	title        = {{CHURRO}: Making History Readable with an Open-Weight Large Vision-Language Model for High-Accuracy, Low-Cost Historical Text Recognition},
	author       = {Semnani, Sina J. and Zhang, Han and He, Xinyan and Tekg{"u}rler, Merve and Lam, Monica S.},
	booktitle    = {Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP 2025)},
	year         = {2025}
}

License

• Code: Apache 2.0
• Model weights: Qwen research license
• Dataset: research use only because of the underlying source licenses