@CoderDaMing: China has open-sourced a peanut-sized OCR that can parse an entire 100-page PDF in one go. It's called 'Unlimited-OCR'. Only 3B parameters. Runs locally. Other OCR tools cut documents page by page, easily losing context. This one reads the entire document at once. → Single 'long-range' parse (32K context window...

X AI KOLs Timeline Models

Summary

China has open-sourced the OCR model Unlimited-OCR with only 3B parameters, which can parse an entire 100-page PDF in one go, supports local execution, achieves 93% accuracy, and is completely free and open-source.

China has open-sourced a peanut-sized OCR that can parse an entire 100-page PDF in one go. It's called 'Unlimited-OCR'. Just 3B parameters. Runs locally. Other OCR tools cut documents page by page, easily losing context. This one reads the entire document at once. → Single 'long-range' parse (32K context window) → Out-of-the-box multilingual support → 93% accuracy on standard benchmarks (+6 above baseline) → Error rate <0.11 on documents over 40 pages → 100% local execution → Supports Transformers, vLLM, SGLang, Docker, Ollama, llama.cpp Traditional cloud OCR (Textract, Google Vision, Azure) costs $1.5–$15 per thousand pages. This runs on your machine. Free forever. Baidu developed it to advance 'DeepSeek-OCR'. It has received 1.9 million downloads on HuggingFace, and most people still don't know about it. 100% open source.
Original Article
View Cached Full Text

Cached at: 07/20/26, 05:31 PM

China has open-sourced a peanut-sized OCR that can parse an entire 100-page PDF at once.

It’s called ‘Unlimited-OCR’. Only 3B parameters. Runs locally.

Other OCR tools split documents page by page, easily losing context. This one reads the entire document in one go.

→ Single ‘long-vision’ parsing (32K context window)
→ Out-of-the-box multilingual support
→ 93% accuracy on standard benchmarks (+6 above baseline)
→ <0.11 error rate on 40+ pages
→ 100% local execution
→ Supports Transformers, vLLM, SGLang, Docker, Ollama, llama.cpp

Traditional cloud OCR (Textract, Google Vision, Azure) costs 1.5–15 per thousand pages.

This runs on your machine. Free forever.

Developed by Baidu to advance ‘DeepSeek-OCR’. Already 1.9 million downloads on Hugging Face, but most people still don’t know about it.

100% open-source.

Similar Articles

@GoSailGlobal: Current OCR processes multi-page documents page by page. Every time you turn a page, memory is reset. Today, Baidu quietly open-sourced a model on GitHub and HuggingFace called Unlimited OCR, inspired by how humans copy books: - When copying a book, you don't reread hundreds of pages every time you write a word...

X AI KOLs Timeline

Baidu has open-sourced the Unlimited OCR model, which uses a Reference Sliding Window Attention (R-SWA) mechanism to parse documents up to 32K context in a single pass, eliminating the need for page-by-page inference.

@geekbb: Baidu's open-source visual language model OCR project, upgraded from DeepSeek-OCR, focuses on one-shot parsing of extremely long documents. The model has two inference modes: 'gundam' mode for dense text in a single image, and 'base' mode for multi-page or PDF processing. https://github…

X AI KOLs Timeline

Baidu has open-sourced the visual language model Unlimited-OCR, upgraded from DeepSeek-OCR, supporting one-shot parsing of extremely long documents, offering two inference modes: gundam (dense text in a single image) and base (multi-page/PDF).

baidu/Unlimited-OCR

Hugging Face Models Trending

Baidu releases Unlimited-OCR, a new model for one-shot long-horizon document parsing, building on Deepseek-OCR. It supports single image and multi-page/PDF parsing via Hugging Face Transformers and SGLang.