⬢github TypeScript · 838 ★ +46 since we first saw it · pushed 12 h ago · Apache-2.0
OpenBMB/MiniCPM-OCR
Fast, accurate OCR with correctness assessment.
MiniCPM-OCR is a family of open-source models for document parsing and text recognition. MiniCPM-OCR 1.0 parses Chinese and English documents and PDFs, including scans, photos, curved pages, and recaptured documents, into text, tables, and formulas. MiniCPM-OCR Jev assesses OCR output at the region level, flagging areas that may need review.
Why now: The models were just open-sourced (Oct 27, 2026) and a Nov 2 update brought on-device iOS document parsing, alongside new SDK releases and an open-source web demo.
Who it is for: Developers building document-processing pipelines who need accurate OCR of Chinese or English documents, with the option to self-host or run locally.
image2textminicpm-vocr
Stars over our 165 snapshots: 792 to 838, since 2 d ago.
Where people talked about it
- ⬢github new repos, most starred 2 d ago
API: https://socialmediatrends-api.osmike.com/v1/repos/OpenBMB/MiniCPM-OCR