Introducing Mistral OCR 3
Get smart on it
Mistral OCR 3 is an optical character recognition model designed to extract text and images from documents and convert them into readable formats like markdown or structured JSON. The model achieves a 74% win rate improvement over its predecessor on forms, scanned documents, tables, and handwriting, with particular enhancements in handling handwritten content, low-quality scans, and complex table structures. It is available through an API and a drag-and-drop interface called Document AI Playground, priced at $2 per 1,000 pages with discounts available for batch processing. The model matters because organizations that can efficiently extract text and images from documents with high accuracy can unlock value from their data and gain competitive advantage in document-based workflows like invoice processing and archive digitization.
Related news
Vibe coding platform Base44 launches own model as AI startups seek defensibility
Wix-owned vibe coding platform Base44 has started rolling out its own AI model — with hopes that it will eventually outperform frontier models.
DiScoFormer: One transformer for density and score, across distributions
DiScoFormer is a transformer model that estimates both the density and score of a data distribution in a single forward pass without requiring retraining for new distributions. The density describes where data points cluster, while the score, the gradient of log-density, points toward more probable regions and is used in diffusion-based generative models, Bayesian sampling, and scientific simulations. Existing methods force a trade-off: classical kernel density estimation works on any distribution but loses accuracy in high dimensions, while neural score-matching models stay accurate in high dimensions but must be retrained for each new distribution. DiScoFormer significantly outperforms kernel density estimation, cutting score error by 6.5 times and density error by more than 37 times in 100 dimensions, while generalizing to distributions with shapes and complexities not seen during training.
