Model · Mistral AI · Vision
Mistral OCR
Mistral's document-understanding model. Specialised OCR — extracts structured text, tables, and layout from PDFs and scans with strong multilingual support.
- Modality
- Vision
- License
- Proprietary (Proprietary)
- Released
- March 6, 2025
- Last verified
- May 10, 2026
- Runs locally
- No
Strengths
- Layout-aware extraction (tables, columns, math)
- Multilingual
- EU-jurisdiction provider
Weaknesses
- Closed weights
- Single-purpose — not a general vision model
Try it
| Where | Type | Notes |
|---|---|---|
| Mistral La Plateforme | hosted-api | API key required |
| Le Chat | product | Document mode |
Used in solutions
Official sources
- OCR docs docs
- Announcement announcement
Change log
- — Initial entry.
- — Marked deprecated after verification that OCR 3 is the current Mistral OCR offering.
Esc