Baidu Open-Sources Unlimited OCR for Long Document Parsing

Baidu has open-sourced Unlimited OCR, a 3B MoE model that parses 40+ page documents with fixed memory via R-SWA. The MIT-licensed repo already has 27k stars, with vLLM support and an arXiv paper.

2026-10-12 ~ 2026-10-12 · 2 related posts