dots.ocr vs Pixtral
Pick dots.ocr for new document parsing work: it is built for PDFs to Markdown with layout, tables as HTML and formulas as LaTeX, and it is MIT-licensed. Pick Pixtral only if you already run on the Mistral API or need question answering over charts and documents, because Mistral marks it deprecated and no longer maintained.
dots.ocrrednote/dots-ocr | Pixtralmistral/pixtral | |
|---|---|---|
| Type | Open source | Vision model |
| Price | 2 credits / page | 2 credits / page |
| Per 1,000 pages | $2.00 | $2.00 |
| Typical speed | 1.5 s / page | 3.0 s / page |
| Best for | Multilingual document parsing | Questions about document images |
| Inputs | PDF, images | images |
| Languages | 100+ languages | Multilingual |
| Max pages per request | 500 | 500 |
| Tables | ✓ Yes | No |
| Handwriting | No | No |
| Math | ✓ Yes | No |
| Bounding boxes | ✓ Yes | No |
| 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 dots.ocr if
- It returns layout boxes, categories and text in one pass, with tables as HTML and formulas as LaTeX.
- It is a 3B model with MIT-licensed code, and the original scores 79.1 on olmOCR-bench (83.9 for the dots.mocr 1.5 release).
- It targets multilingual PDFs with mixed tables and text, and the dots.mocr-svg variant converts charts and diagrams to SVG.
Choose Pixtral if
- It suits question answering over charts and documents, with 90.7 ANLS on DocVQA and 81.8 on ChartQA per Mistral.
- It is available through the Mistral API at a launch price of $0.15 per million input and output tokens, so existing Mistral integrations keep working.
- It has Apache-2.0 open weights and a 128K token context, which allows many images per request.
Key differences
| dots.ocr | Pixtral | |
|---|---|---|
| Purpose | Document parsing model that outputs layout, text, tables and formulas | General vision language model with OCR as one of its abilities |
| Licence | MIT (the repository also includes a separate dots.ocr license agreement file) | Apache-2.0 |
| How you run it | Self-host only, with vLLM or Hugging Face Transformers; single CUDA GPU is the documented path | Mistral API (pixtral-12b-2409) or self-hosted open weights |
| Price | Free to self-host, you pay for compute | $0.15 per million input and output tokens at API launch; free to self-host |
| Size | 3B parameters (1.7B LLM backbone) | 12.4B parameters (12B decoder + 400M vision encoder) |
| Output formats | JSON, Markdown, HTML, LaTeX, bounding boxes, SVG | Plain text, Markdown, JSON |
| Known failure | Runs of ellipses or underscores can make output repeat endlessly; tables above 1000 grid cells raised an error and dropped the page in a Docling issue | Reported accuracy drop on handwriting, small fonts and non-Latin scripts; Q4 quantization reported to degrade small text OCR; deprecated by Mistral |
Sources
- dots.ocr GitHub README
- Docling issue #5146
- Mistral announcement
- Mistral Sept 2024 pricing
- Third-party deployment guide
What each one is
dots.ocr
Single vision model from rednote that does layout detection and content recognition together.
dots.ocr detailsPixtral
Mistral's open-weight vision models, Pixtral 12B and Pixtral Large. Mistral's API no longer serves them, so they run self-hosted.
Pixtral 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": "rednote/dots-ocr","fallbacks": ["mistral/pixtral"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
dots.ocr alternativesPixtral alternativesBest OCR models in 2026