{"name":"MiniCPM-OCR","repo":"OpenBMB/MiniCPM-OCR","url":"https://midorreal.com/project/openbmb-minicpm-ocr","repository":"https://github.com/OpenBMB/MiniCPM-OCR","description":"Fast, accurate OCR with correctness assessment.","facts":{"license":{"value":"Apache-2.0","source":"https://github.com/OpenBMB/MiniCPM-OCR","checked":"22 Nov 2026"},"language":{"value":"TypeScript","source":"https://github.com/OpenBMB/MiniCPM-OCR","checked":"22 Nov 2026"},"last_release":{"value":"sdk-v1.0.5, 9 Nov 2026","source":"https://github.com/OpenBMB/MiniCPM-OCR/releases","checked":"22 Nov 2026"},"releases":{"value":"4","source":"https://github.com/OpenBMB/MiniCPM-OCR/releases","checked":"20 Nov 2026"},"contributors":{"value":"13","source":"https://github.com/OpenBMB/MiniCPM-OCR/graphs/contributors","checked":"20 Nov 2026"},"archived":{"value":"no","source":"https://github.com/OpenBMB/MiniCPM-OCR","checked":"22 Nov 2026"}},"stars":1010.0,"writeup":{"what_it_is":"MiniCPM-OCR is a family of models for document parsing and text recognition. MiniCPM-OCR 1.0 parses Chinese and English documents and PDFs into text, tables, and formulas. MiniCPM-OCR Jev scores OCR results by region to flag content that may need review.","audience":"Developers and users who need to parse or check OCR output from documents and PDFs","claims":[{"kind":"specific","claim":"Achieves a SOTA score of 93.68 on Real5-OmniDocBench","excerpt":"achieves a **SOTA score of 93.68 on Real5-OmniDocBench**","status":"not_checked"},{"kind":"specific","claim":"Achieves a SOTA score of 54.7 on OCRBench v2","excerpt":"achieves a **SOTA score of 54.7 on OCRBench v2**","status":"not_checked"},{"kind":"specific","claim":"Improves end-to-end processing efficiency by 55% with DSpark decoding acceleration","excerpt":"improves **end-to-end processing efficiency by 55%**","status":"not_checked"},{"kind":"specific","claim":"Jev has an AUROC of 93.6 on JevOCRBench with 2B parameters","excerpt":"achieves an AUROC (×100) of 93.6 on [JevOCRBench]","status":"not_checked"},{"kind":"specific","claim":"Supports vLLM, SGLang, llama.cpp, and Ollama from day one","excerpt":"supports **vLLM, SGLang, llama.cpp, and Ollama from day one**","status":"not_checked"}],"alternatives":[],"written_by":"AI, from the project's README","date":"2026-11-14"},"why_now":null,"owner_supplied":null,"score":{"verdict":"real","hype":62,"reality":73,"gap":-11,"momentum":null,"confidence":60,"version":"score-2.0","rule":"verdict-2.0","calculated_at":"2026-11-22T08:00:00+00:00","verdict_source":"rule"},"early_read":null,"changes":[{"date":"2026-11-09T09:42:30+00:00","text":"Latest release sdk-v1.0.5","source":"https://github.com/OpenBMB/MiniCPM-OCR/releases"},{"date":"2026-10-28T10:05:08+00:00","text":"First release","source":"https://github.com/OpenBMB/MiniCPM-OCR/releases"},{"date":"2026-10-27T13:22:00+00:00","text":"Repository created","source":"https://github.com/OpenBMB/MiniCPM-OCR"}],"cite":"MiniCPM-OCR is a GitHub project at https://github.com/OpenBMB/MiniCPM-OCR written mainly in TypeScript, described there as \"Fast, accurate OCR with correctness assessment\".","attribution":"Data from Mid or Real, https://midorreal.com/project/openbmb-minicpm-ocr, checked 22 November 2026. Attribution with a link is required.","still_real":null}