OCRRouter

RapidOCR vs docTR

Pick RapidOCR for offline CPU OCR of Chinese and English screenshots and receipts, especially when you need to embed it in Java, C# or C++ apps. Pick docTR when you want word-level output with coordinates in a nested JSON structure, or when your Python team wants to choose architectures or train its own models. Neither returns tables or Markdown.

RapidOCRrapidai/rapidocrdocTRmindee/doctr
TypeOpen sourceOpen source
Price1 credit / page1 credit / page
Per 1,000 pages$1.00$1.00
Typical speed0.6 s / page0.9 s / page
Best forLightweight OCR on CPU or edgeWord-level OCR with positions
InputsimagesPDF, images
LanguagesChinese, English and moreMainly Latin scripts
Max pages per request100500
TablesNoNo
HandwritingNoNo
MathNoNo
Bounding boxes✓ Yes✓ Yes
Confidence✓ Yes✓ Yes
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

  • It installs with pip install rapidocr onnxruntime and runs PaddleOCR models without the Paddle framework.
  • It has bindings for Python, C++, Java and C#, so it fits desktop and server apps outside Python.
  • It can run on ONNX Runtime, TensorRT, OpenVINO or MNN, and ONNX Runtime on CPU is the default.

Choose docTR if

  • It exports a nested page, block, line and word structure with coordinates as JSON, which suits custom extraction on invoices and receipts.
  • You can choose the detection and recognition architectures and use straight or rotated boxes.
  • It ships ready Docker images and a demo on Hugging Face Spaces, and it includes a key information extraction predictor.

Key differences

RapidOCRdocTR
LicenceApache-2.0. Converted models keep their upstream licences.Apache-2.0.
How you run itLibrary for Python, C++, Java or C#. CPU through ONNX Runtime by default.Python library on PyTorch, with Docker images. CPU or NVIDIA GPU; GPU images need CUDA 12.2 or newer on the host.
Output formatsPlain text and bounding boxes.JSON (page, block, line, word with coordinates), plain text and bounding boxes.
LanguagesChinese and English only by default. Other languages need extra models.Not published
Tables and MarkdownNot for layout, tables or Markdown output.No table, formula or Markdown output. It returns words and lines only.
Known failure16-bit grayscale PNG or TIFF give no text or wrong text (issue #776). Arabic recognition crashes because python-bidi is undeclared (issue #719).Split-crop merging misjoins readings across repeated characters (issue #2229). Cropped images give poorer results (issue #2007).
Export and large pagesSliced inference is unreachable and crashes on large pages (issue #764).ONNX export of MASTER truncates decoding steps (issue #2224).

Sources

What each one is

RapidOCR

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

RapidOCR details

docTR

Open-source OCR library from Mindee with separate text detection and recognition models.

docTR 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": ["mindee/doctr"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

RapidOCR alternativesdocTR alternativesBest OCR models in 2026