OpenBMB open-sources MiniCPM-OCR: 1B model tops OCR leaderboards with a guardrail model
vanstriendaniel · x · 2026-10-28
OpenBMB released the open-source MiniCPM-OCR suite. MiniCPM-OCR 1.0 (1B) delivers SOTA parsing at its scale — 93.68 on Real5-OmniDocBench and 54.7 on OCRBench v2 — beating Xiaomi-OCR-0, PaddleOCR-VL-1.6, GLM-OCR and MinerU2.5-Pro on average scores, running up to 128 pages/min with DSpark. MiniCPM-OCR Jev (2B) is a model-agnostic decision model that region-level checks OCR output, scoring 93.6 on JevOCRBench — above JEV-35B-A3B (90.1) and near Qwen3.5-397B-A17B (93.8). Apache-2.0 on GitHub.
Related event: OpenBMB Open-Sources MiniCPM-OCR Suite, 1.3B Model Hits OCR SOTA(7 posts)→
More from Models
- European 'sovereign AI' startup Giotto bans benchmarking in its ToS, drawing flak — teortaxesTex · 2026-10-28
- Sentdex names the models he's excited about: next Zhipu GLM Flash and Arcee AI's upcoming release — sloppenheimer · 2026-10-28
- How did Anthropic's Opus 5.5 get so good at 3D and game programming? — mark_k · 2026-10-28
- Warning: Letting Grok's Cloud Browser Log Into Your Instagram/Facebook May Get You Banned — gaganghotra_ · 2026-10-28
- New meme: distilled models are now "moonshine," not base whiskey — wavefnx · 2026-10-28
- LiquidAI's LFM3-1.7B edge text-generation model trends on Hugging Face — LiquidAI · 2026-10-28