OCRRouter

Nanonets-OCR vs Nougat

Pick Nougat for English arXiv-style papers with heavy math, but only for non-commercial work, because its weights are CC-BY-NC. Pick Nanonets-OCR for business documents with signatures, watermarks and checkboxes, or for flowcharts you want as Mermaid. Neither is built for handwriting or non-Latin scripts, and Nanonets-OCR's licence is not stated on its model card.

Nanonets-OCRnanonets/nanonets-ocrNougatmeta/nougat
TypeOpen sourceOpen source
Price2 credits / page2 credits / page
Per 1,000 pages$2.00$2.00
Typical speed1.8 s / page3.0 s / page
Best forMarkdown with semantic tagsarXiv-style papers
InputsPDF, imagesPDF, images
LanguagesMultilingualMainly English
Max pages per request500200
Tables✓ Yes✓ Yes
HandwritingNoNo
Math✓ Yes✓ Yes
Bounding boxesNoNo
ConfidenceNoNo
Forms✓ YesNo
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 Nanonets-OCR if

  • Your documents are business PDFs with signatures, watermarks and checkboxes, which it tags or converts to unicode symbols.
  • You need flowcharts or org charts extracted as Mermaid code, or output as HTML as well as Markdown and LaTeX.
  • You want printed PDFs turned into Markdown for LLMs and can accept a 3B model with an olmOCR-bench overall score of 69.5.

Choose Nougat if

  • Your input is English scientific papers, since it was trained on arXiv and PubMed Central documents.
  • You need math-heavy pages as LaTeX, with equations and tables in Mathpix-compatible .mmd Markdown.
  • You want a simple CLI and API with small and base checkpoints, and your use is non-commercial.

Key differences

Nanonets-OCRNougat
Built forBusiness documents with signatures, watermarks, checkboxes and flowchartsScientific PDF pages from arXiv-style papers
LicenceNot stated on the Nanonets-OCR2-3B model card (no licence tag)Code MIT; weights CC-BY-NC, so no commercial use
How you run itSelf-host the weights from Hugging Face, fine-tuned from Qwen2.5-VL-3BSelf-hosted Python package with a CLI and API
Output formatsMarkdown, HTML, LaTeX, MermaidMarkdown (.mmd) and LaTeX
BenchmarksolmOCR-bench overall 69.5, DocVQA 89.43, ChartQA 78.56 (OCR2-3B, reported by the authors)Not published
Languages and handwritingNot trained on handwriting and can hallucinate; handwriting test saw generation loops with CER up to 2100%Works best on English; the README says Chinese, Russian and Japanese do not work
Known failureTrained on a fixed prompt; other prompts are expected to cause hallucinationRepetition loops on 1.5% of test pages, more on out-of-domain documents; looping pages can be dropped as MISSING_PAGE

Sources

What each one is

Nanonets-OCR

Open image-to-Markdown model that tags tables, equations, signatures, watermarks and checkboxes.

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

Questions

More comparisons

Nanonets-OCR alternativesNougat alternativesBest OCR models in 2026