OCRRouter

dots.ocr vs Nougat

Pick dots.ocr for multilingual PDFs, business documents and anything you ship commercially. Pick Nougat only for English scientific papers where you need LaTeX, and only for non-commercial use, because its weights are CC-BY-NC.

dots.ocrrednote/dots-ocrNougatmeta/nougat
TypeOpen sourceOpen source
Price2 credits / page2 credits / page
Per 1,000 pages$2.00$2.00
Typical speed1.5 s / page3.0 s / page
Best forMultilingual document parsingarXiv-style papers
InputsPDF, imagesPDF, images
Languages100+ languagesMainly English
Max pages per request500200
Tables✓ Yes✓ Yes
HandwritingNoNo
Math✓ Yes✓ Yes
Bounding boxes✓ YesNo
ConfidenceNoNo
FormsNoNo
Multilingual✓ YesNo
On OCRRouterAvailableAvailable

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

  • Its code is MIT-licensed, so you can use it in a commercial product (the repo also includes a separate dots.ocr license agreement file, so read it).
  • It handles multilingual PDFs and mixed business and academic documents, and returns layout boxes, categories and text in one pass.
  • It outputs JSON, Markdown, HTML, LaTeX, bounding boxes and SVG, with tables as HTML and formulas as LaTeX.

Choose Nougat if

  • It is built for English arXiv-style papers and was trained on arXiv and PubMed Central documents.
  • It turns math-heavy PDF pages into Mathpix-compatible .mmd Markdown with LaTeX for equations and tables.
  • It ships small and base checkpoints with a simple CLI and API, which makes it easy to try on a folder of papers.

Key differences

dots.ocrNougat
LicenceMIT code, plus a separate dots.ocr license agreement file in the repositoryMIT code, CC-BY-NC weights, so no commercial use
Model size and hardware3B parameters (1.7B LLM backbone). Single CUDA GPU with vLLM is the documented path. No VRAM figure stated.Not published
Output formatsJSON, Markdown, HTML, LaTeX, bounding boxes, SVGMarkdown (.mmd) and LaTeX
LanguagesMultilingualWorks best on English. The README says Chinese, Russian and Japanese do not work.
Document typesMultilingual PDFs, academic and business documents with mixed tables and text, charts to SVG with dots.mocr-svgEnglish scientific papers and math-heavy PDFs. Not suited to business documents, forms or invoices.
BenchmarkolmOCR-bench 79.1 for dots.ocr and 83.9 for dots.mocr (1.5), as reported by the authorsNot published
Known failureRuns of ellipses or underscores can make output repeat endlessly. Complex tables and formulas remain difficult. A table above 1000 grid cells raises an error and drops the page in Docling.Repetition loops on 1.5% of test pages in the paper, more on out-of-domain documents. Repetition detection can drop pages as MISSING_PAGE.

Sources

What each one is

dots.ocr

Single vision model from rednote that does layout detection and content recognition together.

dots.ocr details

Nougat

Meta's model that converts academic PDFs to Markdown with LaTeX math.

Nougat details

Try both on your own file

Run the file once with each model and compare the output.

PDF, PNG, JPG, WebP, HEIC. Up to 20 MB each, 10 files.No file?

Use one, fall back to the other

{"model": "rednote/dots-ocr","fallbacks": ["meta/nougat"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

dots.ocr alternativesNougat alternativesBest OCR models in 2026