OCRRouter

Jina OCR vs Nougat

Pick Jina OCR for general page-to-Markdown work, including a hosted API option. Pick Nougat for English arXiv-style papers where equations must come out as LaTeX. Both have non-commercial weights, so neither can go into a commercial product without a license deal (Jina) or a different model (Nougat).

Jina OCRjina/jina-ocrNougatmeta/nougat
TypeOpen sourceOpen source
Price1 credit / page2 credits / page
Per 1,000 pages$1.00$2.00
Typical speed1.5 s / page3.0 s / page
Best forClean Markdown at low costarXiv-style papers
InputsPDF, imagesPDF, images
LanguagesAbout 100 languagesMainly English
Max pages per request200200
Tables✓ Yes✓ Yes
HandwritingNoNo
Math✓ Yes✓ Yes
Bounding boxesNoNo
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 Jina OCR if

  • It scores 91.14 on OmniDocBench v1.6 and 83.4 on olmOCR-Bench (authors' figures), while Nougat has no published benchmark scores.
  • You can self-host it or call Jina's hosted API, so you can start without running a GPU.
  • It outputs Markdown, HTML and LaTeX, and it has only about 570M active parameters, running at 2.57 pages per second on an A100.

Choose Nougat if

  • It was trained on arXiv and PubMed Central papers, so it fits English scientific PDFs.
  • It writes LaTeX for equations and tables in Mathpix-compatible .mmd Markdown.
  • It ships small and base checkpoints with a simple CLI and Python API, and the code is MIT licensed.

Key differences

Jina OCRNougat
LicenceCC-BY-NC-4.0; commercial use needs a license from Jina/ElasticCode MIT; weights CC-BY-NC, so no commercial use
How you run itSelf-host or Jina's hosted APISelf-hosted Python package only
Output formatsMarkdown, HTML, LaTeXMarkdown (.mmd), LaTeX
BenchmarksOmniDocBench v1.6 overall 91.14; olmOCR-Bench overall 83.4Not published
Model size and hardware3B MoE, about 570M active; GPU recommended, falls back to CPU in TransformersNot published
LanguagesNot publishedWorks best on English; Chinese, Russian and Japanese do not work
Known failureHosted API returns HTTP 503 on cold start; retry after 30 to 60 secondsRepetition loops on 1.5% of test pages in the paper, more on out-of-domain documents; looping pages can be dropped as MISSING_PAGE

Sources

What each one is

Jina OCR

Compact page-to-Markdown model from Jina AI, built on the DeepSeek-OCR encoder.

Jina 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": "jina/jina-ocr","fallbacks": ["meta/nougat"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

Jina OCR alternativesNougat alternativesBest OCR models in 2026