dots.ocr vs Nanonets-OCR
Pick dots.ocr for multilingual PDFs with tables and formulas where you want layout boxes and a permissive MIT-licensed codebase. Pick Nanonets-OCR for printed business documents with signatures, watermarks, checkboxes or flowcharts that you want as Markdown for an LLM. dots.ocr scores higher on olmOCR-bench (79.1 vs 69.5, and 83.9 for dots.mocr), and Nanonets-OCR shows no licence on its model card.
dots.ocrrednote/dots-ocr | Nanonets-OCRnanonets/nanonets-ocr | |
|---|---|---|
| Type | Open source | Open source |
| Price | 2 credits / page | 2 credits / page |
| Per 1,000 pages | $2.00 | $2.00 |
| Typical speed | 1.5 s / page | 1.8 s / page |
| Best for | Multilingual document parsing | Markdown with semantic tags |
| Inputs | PDF, images | PDF, images |
| Languages | 100+ languages | Multilingual |
| Max pages per request | 500 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | No | No |
| Math | ✓ Yes | ✓ Yes |
| Bounding boxes | ✓ Yes | No |
| Confidence | No | No |
| Forms | No | ✓ Yes |
| 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 scores 79.1 on olmOCR-bench, against 69.5 for Nanonets-OCR2-3B, and the newer dots.mocr reaches 83.9.
- One pass returns layout boxes, categories and text, with tables as HTML and formulas as LaTeX.
- The code is MIT-licensed, which makes the licence question simpler than for Nanonets-OCR.
Choose Nanonets-OCR if
- It tags signatures, watermarks and image descriptions, and turns checkboxes into unicode symbols.
- It extracts flowcharts and org charts as Mermaid code.
- It scores 89.43 on DocVQA for the 3B model, and it separates inline from display LaTeX equations.
Key differences
| dots.ocr | Nanonets-OCR | |
|---|---|---|
| Licence | MIT (the repository also has a separate dots.ocr license agreement file) | Not stated on the Nanonets-OCR2-3B model card; base model is Qwen2.5-VL-3B |
| How you run it | Self-hosted with vLLM or Hugging Face Transformers; single CUDA GPU with vLLM is the documented path | Self-hosted from Hugging Face weights; both are free apart from compute |
| Output formats | JSON, Markdown, HTML, LaTeX, bounding boxes, SVG | Markdown, HTML, LaTeX, Mermaid |
| olmOCR-bench | 79.1 (dots.ocr), 83.9 (dots.mocr), reported by the authors | 69.5 overall (OCR2-3B), reported by the authors |
| Special elements | Layout boxes and categories; charts and diagrams to SVG with dots.mocr-svg | Signatures, watermarks, checkboxes, flowcharts to Mermaid |
| Prompting | Not published | Trained on a fixed prompt; other prompts are expected to cause hallucination |
| 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 | Not trained on handwriting and can hallucinate; a handwriting test saw generation loops with CER up to 2100% |
Sources
- dots.ocr GitHub README
- dots.ocr blog / limitations
- Docling issue #5146
- Nanonets-OCR2-3B model card
- Nanonets-OCR-s announcement
- HF discussion fixed prompt
What each one is
dots.ocr
Single vision model from rednote that does layout detection and content recognition together.
dots.ocr detailsNanonets-OCR
Open image-to-Markdown model that tags tables, equations, signatures, watermarks and checkboxes.
Nanonets-OCR 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": ["nanonets/nanonets-ocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
dots.ocr alternativesNanonets-OCR alternativesBest OCR models in 2026