Understand this story
On October 28, 2026, ModelBest OpenBMB launched the open-source MiniCPM-OCR suite, comprising the 1B parsing model MiniCPM-OCR 1.0 and the 2B decision model MiniCPM-OCR Jev. According to information released by ModelBest OpenBMB, MiniCPM-OCR 1.0 scored 93.68 on Real5-OmniDocBench and 54.7 on OCRBench v2; Jev can cross-validate text, formulas, and tables across models, scoring 93.6 on JevOCRBench, surpassing JEV-35B-A3B's 90.1 and approaching Qwen3.5-397B-A17B's 93.8. With DSpark, its speed is 1.49 times that of PaddleOCR-VL-1.6 and 3.62 times that of MinerU2.5 Pro; after SDK concurrency optimization, it can reach 128 pages per minute, and it supports deployment on vLLM, SGLang, llama.cpp, and Ollama. Another report on the same day stated that MiniCPM-OCR 1.0 is a 1.3B-parameter model that supports Chinese and English documents, can handle scanned pages, photographed documents, and skewed and curved pages, and supports tables, formulas, PDFs, and key information extraction; Jev is used to evaluate the reliability of OCR results and help identify outputs that require manual review. Both models have released weights and offer multiple deployment methods.
AI-generated from public reports
Latest update · Discussion
English text is being prepared. joined this discussionDevelopments
1 confirmed developments
- Source:English text is being prepared.
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Oct 29
English text is being prepared.Related discussion
English text is being prepared. joined this discussion
This post contributes to heat; it is not a formal report or confirmed fact.
Oct 28
OpenBMB
OpenBMB Releases MiniCPM-OCR 1.0 and Verification Model Jev
OpenBMB releases MiniCPM-OCR 1.0, a 1B-scale OCR model that achieves 93.68 on Real5-OmniDocBench and 54.7 on OCRBench v2. It also releases MiniCPM-OCR Jev, a 2B decision model used to compare parsing results with the original image and give a 0–1 correctness score, approaching Qwen3.5-397B-A17B on JevOCRBench. Both models have open weights, support deployment via llama.cpp, vLLM, SGLang, and Ollama, and come with open-source fine-tuning code.
ModelBest OpenBMB
ModelBest OpenBMB Launches the MiniCPM-OCR Model Family, Including Two Models for Parsing and Reliability Assessment
ModelBest OpenBMB and ModelBest have launched the open-source model family MiniCPM-OCR, designed to convert documents into structured information. Among them, MiniCPM-OCR 1.0 is a 1.3B-parameter model that supports Chinese and English documents, can handle scanned pages, photographed documents, tilted and curved pages, and supports tables, formulas, PDFs, and key information extraction; the other model, MiniCPM-OCR Jev, is used to assess the reliability of OCR results and help identify outputs that require manual review. Both models have released their weights and offer multiple deployment methods.
Paul CouvertRelated discussion
Paul Couvert joined this discussion
This post contributes to heat; it is not a formal report or confirmed fact.
OpenBMB
OpenBMB Open-Sources MiniCPM-OCR Suite, Featuring a 1B Parsing Model and a 2B Verification Model
OpenBMB has released the open-source MiniCPM-OCR suite, comprising the 1B MiniCPM-OCR 1.0 and the 2B decision model MiniCPM-OCR Jev. MiniCPM-OCR 1.0 scores 93.68 on Real5-OmniDocBench and 54.7 on OCRBench v2; Jev can perform cross-model verification of text, formulas, and tables, scoring 93.6 on JevOCRBench, surpassing JEV-35B-A3B's 90.1 and approaching Qwen3.5-397B-A17B's 93.8. Powered by DSpark, its speed is 1.49 times that of PaddleOCR-VL-1.6 and 3.62 times that of MinerU2.5 Pro; with SDK concurrency optimization, it can reach 128 pages per minute, and it supports deployment on vLLM, SGLang, llama.cpp, and Ollama.
Heat trend for this story
- Current
- 5
- Period peak
- 23 Oct 28 11:00
- 24-hour change
- —
Move the pointer or select the chart to inspect hourly heat. Use the left and right arrow keys with a keyboard.