OCRRouter

RapidOCR vs Nougat

Pick RapidOCR for offline CPU OCR of Chinese and English screenshots and receipts, or when you need to embed OCR in a Java, C# or C++ app. Pick Nougat only for English scientific papers where you need Markdown with LaTeX math, and only if your use is non-commercial. The main split is output and licence: RapidOCR returns plain text and boxes under Apache-2.0, while Nougat returns structured Markdown but its weights are CC-BY-NC.

RapidOCRrapidai/rapidocrNougatmeta/nougat
TypeOpen sourceOpen source
Price1 credit / page2 credits / page
Per 1,000 pages$1.00$2.00
Typical speed0.6 s / page3.0 s / page
Best forLightweight OCR on CPU or edgearXiv-style papers
InputsimagesPDF, images
LanguagesChinese, English and moreMainly English
Max pages per request100200
TablesNo✓ Yes
HandwritingNoNo
MathNo✓ Yes
Bounding boxes✓ YesNo
Confidence✓ YesNo
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 RapidOCR if

  • You need a commercial-friendly licence: the code is Apache-2.0, though converted models keep their upstream licences.
  • You want to embed OCR in Python, C++, Java or C# apps and run it on CPU with ONNX Runtime, or on TensorRT, OpenVINO or MNN.
  • Your input is Chinese or English screenshots and receipts, where plain text and bounding boxes are enough.

Choose Nougat if

  • You process English arXiv-style papers and want equations and tables as LaTeX in Mathpix-compatible Markdown.
  • Your project is non-commercial, since the weights are free under CC-BY-NC.
  • You want a simple CLI and API with small and base checkpoints, trained on arXiv and PubMed Central documents.

Key differences

RapidOCRNougat
LicenceApache-2.0; converted models keep their upstream licencesCode MIT; weights CC-BY-NC, so no commercial use
How you run itLibrary for Python, C++, Java or C#; CPU via ONNX Runtime by defaultSelf-hosted Python package with a CLI and API; small and base checkpoints
OutputPlain text and bounding boxesMarkdown with LaTeX for math and tables
Best document typeChinese and English screenshots and receiptsEnglish arXiv-style scientific papers and math-heavy PDFs
LanguagesChinese and English by default; other languages need extra modelsWorks best on English; Chinese, Russian and Japanese do not work
Known failure16-bit grayscale PNG or TIFF give no text or wrong text; Arabic crashes without manual setupRepetition loops on 1.5% of test pages in the paper, more on out-of-domain documents; looped pages can be dropped as MISSING_PAGE
PriceFree, self-hosted; you pay for computeFree for non-commercial use, self-hosted; you pay for compute

Sources

What each one is

RapidOCR

PaddleOCR models packaged for ONNX Runtime and other engines. Easy to deploy without Paddle.

RapidOCR 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": "rapidai/rapidocr","fallbacks": ["meta/nougat"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

RapidOCR alternativesNougat alternativesBest OCR models in 2026