Mistral AI Launches New Generation Document Recognition Model Mistral OCR 4
Mistral AI has launched its latest document recognition model, Mistral OCR 4, which supports 170 languages across 10 language families and scored 93.07 on the OmniDocBench benchmark. Compared to competitors such as GPT 5.5 Pro and Gemini 3.1 Pro Preview, the model's output is more preferred by humans. In addition to outputting text, the model can provide bounding boxes, region classification, and confidence scores, supporting downstream workloads such as RAG semantic chunking.
Mistral AI has launched its latest document recognition model, Mistral OCR 4, which supports 170 languages across 10 language families and scored 93.07 on the OmniDocBench benchmark. Compared to competitors such as GPT 5.5 Pro and Gemini 3.1 Pro Preview, the model's output is more preferred by humans. In addition to outputting text, the model can provide bounding boxes, region classification, and confidence scores, supporting downstream workloads such as RAG semantic chunking.
This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.