- Home
- AI OCR Software
- Mistral OCR
Mistral OCR
Paris, France
What it is
Mistral OCR is a document-understanding and text-extraction service available through API, self-hosted deployment and the web. OCR 4.1 provides paragraph-level bounding boxes, structural block labels and confidence scores for individual blocks. OCR 4 supports 170 languages across 10 language groups, along with table extraction, structure extraction and handwriting OCR. The API supports batch processing, while Document AI in Mistral Studio offers a no-code interface. OCR 4 can run in a single container on an organization’s infrastructure, with self-managed deployment available to enterprise customers. Mistral says self-hosting keeps document data in the customer’s environment for residency, sovereignty and compliance needs. The service is an ingestion component of Mistral Search Toolkit, an open-source search framework for RAG and enterprise search. The OCR 4.1 API costs $4 per 1,000 pages per month, and Document AI costs $5 per 1,000 annotated pages. Python and TypeScript SDKs are listed.
For and against
For
- Supports 170 languages
- Provides tables, structure and handwriting OCR
- Offers API batching
- Can run in a single self-hosted container
- Includes Python and TypeScript SDKs
Against
- Not intended for real-time processing
- Excludes medical, legal and high-stakes decisions
- Maximum file size is 512 MB
- Paid plans begin at $4
The verdict
Mistral OCR combines multilingual extraction with structural output, batch APIs and self-hosting. Its stated limitations make it unsuitable for high-stakes or real-time document uses.
Who it suits
Mistral OCR suits teams extracting structured text from documents, including organizations needing batch, no-code or self-hosted options. Enterprise users can choose self-managed deployment and premium API support.
Plans and prices
OCR 4.1 API
OCR processing · batch API available
Read 29 Sep 2026 · docs.mistral.ai
Document AI
Document AI processing
Read 29 Sep 2026 · docs.mistral.ai
On the record
- Batch processing
- The OCR API supports batching through the Batch API.docs.mistral.ai · read 29 Sep 2026
- Integration
- OCR 4 is an ingestion component of Mistral Search Toolkit, an open-source search framework for RAG and enterprise search.mistral.ai · read 29 Sep 2026
- Languages
- OCR 4 supports 170 languages across 10 language groups.mistral.ai · read 29 Sep 2026
- Limitations
- Mistral says OCR 4 is not intended for medical diagnosis, legal advice or judgment, high-stakes financial decisions, safety-critical systems, real-time processing, or non-document inputs.mistral.ai · read 29 Sep 2026
- Privacy
- Mistral says self-hosting keeps document data in the customer’s environment for residency, sovereignty, and compliance needs.mistral.ai · read 29 Sep 2026
- Purpose
- OCR 4.1 is Mistral’s OCR service for document understanding and text extraction.docs.mistral.ai · read 29 Sep 2026
- Self-hosting
- OCR 4 can run in a single container on an organization’s own infrastructure; self-managed deployment is available to enterprise customers.mistral.ai · read 29 Sep 2026
- Structured extraction
- OCR 4.1 provides paragraph-level bounding boxes, structural block labels, and block-level confidence scores.docs.mistral.ai · read 29 Sep 2026
- Support
- Mistral’s Enterprise APIs include premium support and system-level SLAs.mistral.ai · read 29 Sep 2026
- Ways to use it
- OCR is available through an API, and teams can use Document AI in Mistral Studio as a no-code interface.mistral.ai · read 29 Sep 2026
