OCRRouter

Nougat vs Unstructured

Pick Nougat for English arXiv-style papers when you want LaTeX math and tables out as Markdown, and your use is non-commercial. Pick Unstructured for mixed file types such as PDFs, Word, PowerPoint, emails and HTML, or for any commercial product. The main split is licence and scope: Nougat's weights are CC-BY-NC and it targets scientific papers, while Unstructured has an Apache-2.0 library and a hosted API for general ingestion.

Nougatmeta/nougatUnstructuredunstructured/unstructured
TypeOpen sourceOCR API
Price2 credits / page20 credits / page
Per 1,000 pages$2.00$20.00
Typical speed3.0 s / page3.0 s / page
Best forarXiv-style papersMixed file types into RAG
InputsPDF, imagesPDF, images, DOCX
LanguagesMainly EnglishMultilingual
Max pages per request2001,000
Tables✓ Yes✓ Yes
HandwritingNoNo
Math✓ YesNo
Bounding boxesNo✓ Yes
ConfidenceNoNo
FormsNoNo
MultilingualNo✓ Yes
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 Nougat if

  • Your input is English scientific papers, since it was trained on arXiv and PubMed Central documents.
  • You need equations and tables as LaTeX inside Mathpix-style Markdown.
  • Your use is non-commercial and you can self-host it and pay only for compute.

Choose Unstructured if

  • You ingest many file types, including PDFs, HTML, Word, PowerPoint, emails and images, through one library.
  • You are building a commercial product and need an Apache-2.0 library or a hosted API.
  • You want to try a hosted service first, with 10,000 free pages and then $0.015 per page.

Key differences

NougatUnstructured
LicenceCode MIT; weights CC-BY-NC, so no commercial useApache-2.0 for the open-source library; proprietary for the hosted platform
How you run itSelf-hosted Python package with a CLI and APISelf-host the library or call the hosted API
PriceFree for non-commercial use; you pay for computeLibrary free; hosted gives 10,000 free pages, then $0.015 per page; Business is custom
Output formatsMarkdown and LaTeXJSON, HTML, plain text and bounding boxes
Input typesScientific PDF pagesPDFs, HTML, Word, PowerPoint, emails and images
LanguagesWorks best on English; Chinese, Russian and Japanese do not workTable OCR ignores language settings, so non-English tables are a weak spot
Known failureRepetition loops on 1.5% of test pages in the paper, more on out-of-domain documents; looping pages can be dropped as MISSING_PAGEOpen-source table cell content scores 0.426 versus 0.866 hosted; hi_res tables can lose the first letters at the edges

Sources

What each one is

Nougat

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

Nougat details

Unstructured

Document ETL for RAG. Partitions many file types into typed elements; open-source library too.

Unstructured 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": "meta/nougat","fallbacks": ["unstructured/unstructured"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

Nougat alternativesUnstructured alternativesBest OCR models in 2026