dots.ocr vs Typhoon OCR
Pick Typhoon OCR for Thai and Thai-English documents such as government forms, financial reports and books, and when you want a hosted API option. Pick dots.ocr for multilingual PDFs where you need layout boxes and Markdown from a self-hosted model. The main split is language scope: Typhoon OCR is evaluated only on Thai and English, while dots.ocr is built for multilingual documents.
dots.ocrrednote/dots-ocr | Typhoon OCRtyphoon/typhoon-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 | Thai documents |
| Inputs | PDF, images | PDF, images |
| Languages | 100+ languages | Thai, English |
| Max pages per request | 500 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| 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
- You work with multilingual PDFs and want Markdown with layout, since Typhoon OCR was evaluated only on Thai and English.
- You need bounding boxes, categories and text from one pass, with tables as HTML and formulas as LaTeX, plus JSON output.
- You want to convert charts and diagrams to SVG with the dots.mocr-svg variant.
Choose Typhoon OCR if
- Your documents are Thai or Thai-English, such as government forms, financial reports and books. V1.5 beat GPT-5 and Gemini 2.5 Pro on Thai government forms (BLEU 0.870).
- You want a smaller model: V2 is 2B (with a 0.8B variant) and scores 84.7 on olmOCR-bench, against 79.1 for the original dots.ocr.
- You want the option of a hosted API (typhoon-ocr-v2) as well as self-hosting.
Key differences
| dots.ocr | Typhoon OCR | |
|---|---|---|
| Language focus | Multilingual documents | Thai and English only; no other languages evaluated |
| How you run it | Self-host only, via vLLM or Hugging Face Transformers | Self-host the weights or use the Typhoon API (typhoon-ocr-v2) |
| Licence | MIT, with a separate dots.ocr license agreement file in the repository | Apache-2.0, subject to OpenTyphoon Terms and Conditions |
| Model size | 3B (1.7B LLM backbone) | 2B (V2 and V1.5); 0.8B V2 variant |
| Output formats | JSON, Markdown, HTML, LaTeX, bounding boxes, SVG | Markdown, HTML, LaTeX |
| olmOCR-bench (authors' numbers) | 79.1 (dots.ocr); 83.9 (dots.mocr / 1.5) | 84.7 (V2 2B) |
| Known failure | Runs of ellipses or underscores can make output repeat endlessly; a table above 1000 grid cells raises an error and drops the page in Docling | V2 can loop until the token limit, mostly on dense multi-column pages; key-information extraction is weakest at 21.0 on ThaiOCRBench |
Sources
What each one is
dots.ocr
Single vision model from rednote that does layout detection and content recognition together.
dots.ocr detailsTyphoon OCR
Open vision model tuned for Thai and English documents, with Markdown output.
Typhoon 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": ["typhoon/typhoon-ocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
dots.ocr alternativesTyphoon OCR alternativesBest OCR models in 2026