OCRRouter

Marker vs OCRmyPDF

Pick Marker when you need Markdown, HTML, JSON or RAG chunks from PDFs such as books and papers with equations, and you can run a GPU or CPU yourself. Pick OCRmyPDF when you need to turn scanned PDFs into searchable PDF/A archives and nothing more. The main split is output: Marker extracts structured content, OCRmyPDF only adds a text layer.

Markerdatalab/markerOCRmyPDFocrmypdf/ocrmypdf
TypeOCR APIOpen source
Price14 credits / page1 credit / page
Per 1,000 pages$14.00$1.00
Typical speed2.5 s / page1.0 s / page
Best forBooks and research papersSearchable PDFs from scans
InputsPDF, images, DOCXPDF, images
Languages90+ languages100+ languages (via Tesseract)
Max pages per request1,0001,000
Tables✓ YesNo
HandwritingNoNo
Math✓ YesNo
Bounding boxes✓ YesNo
ConfidenceNoNo
FormsNoNo
Multilingual✓ Yes✓ 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 Marker if

  • You need PDF to Markdown, HTML, JSON block trees or RAG chunks, which OCRmyPDF does not produce.
  • You work with books and papers with equations and want LaTeX output.
  • You want a hosted option: Datalab sells an API with about 5,000 free pages per month.

Choose OCRmyPDF if

  • You need validated searchable PDF/A for archiving or compliance workflows.
  • You want a text layer inserted losslessly without re-encoding the rest of the PDF, on CPU only.
  • You need Tesseract language packs covering 100+ languages, with work spread across all CPU cores.

Key differences

MarkerOCRmyPDF
What it doesConverts PDF to Markdown, built on the Surya OCR modelsAdds a Tesseract OCR text layer to scanned PDFs
Output formatsMarkdown, HTML, JSON, chunks, LaTeX, imagesSearchable PDF/A, plain text
LicenceCode Apache-2.0; weights under a modified AI Pubs Open Rail-M licence, free for research, personal use and startups under $5MMPL-2.0
PriceFree self-hosted within the weight licence, you pay for compute; hosted API has about 5,000 free pages per monthFree, self-hosted, you pay for compute
HardwareGPU, CPU or Apple Silicon; vllm mode needs an NVIDIA GPU; fast mode runs on CPU with about 18 MB of ONNX weightsCPU only, uses all cores
BenchmarkolmOCR-bench 83.7 in vllm mode, 66.6 in CPU fast mode (Datalab's own numbers)Not published
Known failureWeights block free commercial use for companies above $5MOutput can grow 2x to 4x in size with --deskew --clean-final; --skip-text with default PDF/A removes form fields; hangs on PDFs whose forms share one Resources dictionary

Sources

What each one is

Marker

Hosted version of the open-source Marker and Surya models. Strong on books and papers.

Marker details

OCRmyPDF

Command-line tool that adds a searchable text layer to scanned PDFs, using Tesseract.

OCRmyPDF 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": "datalab/marker","fallbacks": ["ocrmypdf/ocrmypdf"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

Marker alternativesOCRmyPDF alternativesBest OCR models in 2026