Model · Mistral AI · Vision
OCR 3
Mistral's previous OCR service for its Document AI stack, extracting interleaved text and images from documents. Superseded by OCR 4.1, but still supported for existing integrations — and half the price.
- Modality
- Vision
- License
- Mistral Premier (Proprietary)
- Released
- December 18, 2025
- Last verified
- September 8, 2026
- Runs locally
- No
Strengths
- Purpose-built for document extraction
- Handles interleaved text and images
- Current OCR model in Mistral's Document AI stack
Weaknesses
- Single-purpose model rather than a general vision model
- Premier / proprietary access
Try it
| Where | Type | Notes |
|---|---|---|
| Mistral La Plateforme | hosted-api | Premier API access |
Version history
- OCR 4.1 Jul 2026
- OCR 3 Dec 2025 Current Deprecated
Official sources
- Model card docs
- Announcement announcement
Change log
- — Superseded by OCR 4.1. Release date corrected to 2025-12-18 to match Mistral's model card. Still available and not on Mistral's deprecation table.
- — Verified current vs Mistral docs. Corrected release date to 2025-12-17 (announcement date).
- — Added after verification of Mistral's current OCR lineup.
Esc