Mistral Introduces OCR API for AI-Ready Document Conversion

Mistral has launched a new Optical Character Recognition (OCR) API designed to convert PDF documents into AI-ready Markdown files, announced on their website. This API aims to facilitate the ingestion of complex documents by AI models, particularly those relying on raw text formats.

Unlike traditional OCR solutions, Mistral OCR is multimodal, capable of recognizing and preserving both text and graphical elements such as illustrations and photos. The output is formatted in Markdown, a syntax widely used by developers for its ability to include links, headers, and other formatting elements.

Mistral OCR is available through Mistral's API platform and cloud partners like AWS, Azure, and Google Cloud Vertex. For organizations handling sensitive data, an on-premise deployment option is also offered. The API is noted for its superior performance compared to existing solutions from Google, Microsoft, and OpenAI, especially with complex documents and non-English texts.

The company has integrated Mistral OCR into its AI assistant, Le Chat, to enhance document processing capabilities. This tool is expected to be particularly useful for industries like law and research, where handling large volumes of documents is common.