A naukri.com initiative
VentureBeat
1M
36
Image Credit: VentureBeat
Mistral releases new optical character recognition (OCR) API claiming top performance globally
- French AI startup Mistral has launched Mistral OCR, an optical character recognition API for advanced document understanding.
- The API extracts content from unstructured PDFs and images with high accuracy, presenting structured data for easier analysis.
- Unstructured data, which lacks a specific format, includes various data types like images, audio files, and social media posts.
- With multilingual support and fast processing speeds, Mistral OCR aims to assist organizations in digitizing and organizing their data.
- The API offers improved document comprehension by interpreting elements beyond text extraction, such as tables and mathematical expressions.
- Mistral OCR outperforms competitors in math recognition, scanned documents, and multilingual text processing.
- The API is integrated into Le Chat for document processing and is available through Mistral's developer suite.
- Mistral OCR's speed advantage allows processing up to 2,000 pages per minute and is suitable for high-volume document workflows.
- The tool offers efficiency, security, and scalability for document-driven workflows in industries like finance, healthcare, and legal.
- Priced at 1,000 pages per $1, Mistral OCR is available for developers to test on Mistral's website and plans to expand to cloud partners.
Read Full Article
2 Likes
For uninterrupted reading, download the app