Mistral AI has introduced Mistral OCR 3, a new version of their optical character recognition model, which boasts a 74% overall win rate over its predecessor, Mistral OCR 2, on various document types, including forms, scanned documents, and handwritten content. The model is designed to extract text and embedded images from a wide range of documents with high fidelity and is available at a price of $2 per 1,000 pages. Mistral OCR 3 is suitable for both high-volume enterprise pipelines and interactive document workflows. It can be accessed through the API or the Document AI Playground interface in Mistral AI Studio.