OpenBMB open-sources MiniCPM-OCR: 1B SOTA OCR plus a 2B accuracy-checker model
alifcoder · x · 2026-10-28
OpenBMB released the open-source MiniCPM-OCR suite pairing fast parsing with result verification:
- MiniCPM-OCR 1.0: a 1B model scoring 93.68 on Real5-OmniDocBench and 54.7 on OCRBench v2, claimed SOTA at its scale, with speeds up to 128 pages per minute.
- MiniCPM-OCR Jev: a 2B, model-agnostic decision model that checks text, formulas and tables from other OCR engines, approaching Qwen3.5-397B-A17B on JevOCRBench.
The suite targets the failure mode where OCR output looks fine but details are wrong in high-stakes workflows.
Related event: OpenBMB Open-Sources MiniCPM-OCR: 1.3B Parser Plus 2B Quality Judge(12 posts)→
More from Models
- Claim: Anthropic Holds Back Stronger Claude Models While OpenAI Struggles — haider1 · 2026-10-28
- LiquidAI's edge-focused 2B vision-language model LFM3-VL-2B trends on Hugging Face — LiquidAI · 2026-10-28
- Dev blogger slams OpenAI: constant reset notices, slower and dumber models — lxfater · 2026-10-28
- Datalab lite OCR costs $0.50 per 1,000 pages, 10x-30x cheaper than Azure, Google, AWS — VikParuchuri · 2026-10-28
- Screenshot Suggests Gemini 4 Argon Is Rolling Out to Pro Subscribers with Weekly Quota — arg-developer · 2026-10-28
- Gemini Notebook reportedly adding Connectors, with Google Drive as the only launch integration — testingcatalog · 2026-10-28