OpenBMB 发布 1.3B 开源 OCR 模型 MiniCPM-OCR 1.0，在 OCRBench v2 上从 47.4 提升到 54.7，CHAOS-Bench 从 21.28 提升到 29.78，吞吐为 PaddleOCR-VL-1.6 的 1.49 倍。
OpenBMB presents MiniCPM-OCR 1.0, a 1.3B open-source OCR model with open-SOTA results across major benchmarks.
- OCRBench v2: 47.4 -> 54.7
- CHAOS-Bench: 21.28 -> 29.78
- Competitive throughput: 1.49× PaddleOCR-VL-1.6
Recently, open OCR models have rapidly improved to the point where they often exceed proprietary OCR products offered by hyperscalers, while costing an order of magnitude less. This release pushes the trend further. Honestly, it makes no sense to keep paying $10 per 1k pages on proprietary products anymore when you can pay <$1 per 1k pages.
Day-one support: vLLM, SGLang, llama.cpp and Ollama, plus quantized versions.
Also released: MiniCPM-OCR Jev, a 2B quality judge for flagging unreliable OCR outputs and prioritizing review. It achieves 93.6 on JevOCRBench, outperforming JEV-35B-A3B (90.1) and approaching Qwen3.5-397B-A17B (93.8).
HF (MiniCPM-OCR 1.0): https://huggingface.co/openbmb/MiniCPM-OCR-1.0
HF (MiniCPM-OCR Jev): https://huggingface.co/openbmb/MiniCPM-OCR-Jev
GitHub: https://github.com/OpenBMB/MiniCPM-OCR
@OpenBMB
Your OCR says it’s done—but did it get everything right? 🤔 In high-stakes workflows, even a small error can be costly. 🚀 Meet the MiniCPM-OCR suite: an open-source solution to parse fast and guard the results. Two new models: 🔹 MiniCPM-OCR 1.0 — SOTA OCR performance and speed at 1B scale. Compact, accurate, and fast. 🔹 MiniCPM-OCR Jev — A 2B decision model that helps assess parsing accuracy, with performance approaching that of Qwen3.5-397B-A17B on JevOCRBench. ⚡ What’s new? Small model, SOTA results: MiniCPM-OCR 1.0 scores 93.68 on Real5-OmniDocBench and 54.7 on OCRBench v2. A sharp goalkeeper 🛡️: MiniCPM-OCR Jev is model-agnostic, checking text, formulas, and tables from different OCR parsers. It achieves 93.6 on JevOCRBench, outperforming JEV-35B-A3B (90.1) and approaching Qwen3.5-397B-A17B (93.8). Up to 128 pages per minute 🤯: With DSpark, MiniCPM-OCR runs at 1.49× the speed of PaddleOCR-VL-1.6 and 3.62× that of MinerU2.5 Pro. The SDK’s concurrency optimizations further boost throughput to as much as 128 pages per minute. Deploy where you work: Day-0 support for Arm, Houmo AI, Intel, and NVIDIA, plus 10 more chip vendors through FlagOS. Ready for your stack: vLLM, SGLang, llama.cpp, and Ollama, with quantized models, fine-tuning support, and Agent Skills. 🔗 GitHub: http://github.com/OpenBMB/MiniCPM-OCR 🤗 Hugging Face: MiniCPM-OCR 1.0：http://huggingface.co/openbmb/MiniCPM-OCR-1.0 MiniCPM-OCR Jev：http://huggingface.co/openbmb/MiniCPM-OCR-Jev 👉 Try the Demo：http://platform.modelbest.ai/console/playground/document-parse Use the API：http://platform.modelbest.ai/console/docs/models/document-parse在 X 查看被引用的帖子
来源：Aran Komatsuzaki · x.com