Mistral OCR vs Chunkr
Pick Mistral OCR for high-volume PDF to Markdown for RAG and batch scans, mainly because of the low list price: $2 per 1,000 pages, $1 with the Batch API. Pick Chunkr when you need self-hosting, bounding boxes or semantic chunking of PDFs and slides. Mistral OCR is a hosted API only, so Chunkr is the choice if documents cannot leave your infrastructure.
Mistral OCRmistral/mistral-ocr | Chunkrchunkr/chunkr | |
|---|---|---|
| Type | OCR API | OCR API |
| Price | 25 credits / page | 14 credits / page |
| Per 1,000 pages | $25.00 | $14.00 |
| Typical speed | 0.9 s / page | 3.0 s / page |
| Best for | High-volume PDF to Markdown | Layout-aware chunks for RAG |
| Inputs | PDF, images | PDF, images, DOCX |
| Languages | 100+ languages | Multilingual |
| Max pages per request | 1,000 | 1,000 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | No | No |
| Math | ✓ Yes | ✓ Yes |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | No | No |
| Forms | No | No |
| Multilingual | ✓ Yes | ✓ Yes |
| On OCRRouter | Available | Available |
Prices are what you pay on OCRRouter. Speed is the provider's typical figure, not a benchmark we ran.
Which to choose
Choose Mistral OCR if
- It costs $2 per 1,000 pages, or $1 per 1,000 through the Batch API.
- OCR 3 rebuilds tables as HTML with rowspan and colspan.
- It suits scanned business documents processed in batch, with Markdown, HTML, JSON and images as output.
Choose Chunkr if
- You can self-host it with Docker, with GPU, CPU-only and Mac ARM options.
- It returns OCR with bounding boxes and does layout analysis and semantic chunking.
- It accepts PDF, PPT, Word and images, and works with any OpenAI-compatible LLM endpoint.
Key differences
| Mistral OCR | Chunkr | |
|---|---|---|
| How you run it | Hosted API only, through the Mistral API (pinned version mistral-ocr-2512, OCR 3). | Cloud API or self-hosted open-source version. |
| Licence | Proprietary. | AGPL-3.0, or a commercial licence. Closed-source use needs the commercial licence. |
| Price | $2 per 1,000 pages; $3 with structured annotations; $1 per 1,000 via Batch API. | Hosted credit-based plans with unverified prices. The open-source version is free, but you pay for compute and an LLM endpoint. |
| Output | Markdown, HTML, JSON and extracted images. | HTML, Markdown, JSON and bounding boxes. |
| Benchmark | OmniDocBench v1.6 overall: 85.66. The first version scored 78.83 on v1.5. | Not published |
| Known failure | Hallucinated text on unreadable handwriting, and changed checkbox states and misread characters on a dense financial table page in Reducto's test. | The open-source version uses community models that are less accurate than the cloud's in-house models. It does not support Excel. |
| Input formats | Scanned business documents and PDFs. | PDF, PPT, Word and images. Excel is not supported in the open-source version. |
Sources
- Mistral OCR 3 announcement
- VentureBeat on OCR 3 pricing
- OmniDocBench v1.6 leaderboard
- Reducto test of Mistral OCR
- GitHub README
- Chunkr site
What each one is
Mistral OCR
Fast document OCR that returns clean Markdown with tables and embedded images.
Mistral OCR detailsChunkr
Document parsing API that segments and chunks pages for RAG. Self-hostable version available.
Chunkr detailsTry both on your own file
Run the file once with each model and compare the output.
Use one, fall back to the other
{"model": "mistral/mistral-ocr","fallbacks": ["chunkr/chunkr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Mistral OCR alternativesChunkr alternativesBest OCR models in 2026