OCRRouter

OCRmyPDF vs Nanonets-OCR

Pick OCRmyPDF to make scanned PDFs searchable and keep them as PDF/A. Pick Nanonets-OCR to turn printed business documents into Markdown for an LLM pipeline. They solve different problems: OCRmyPDF adds a text layer to a PDF, Nanonets-OCR reads the page and writes structured text.

OCRmyPDFocrmypdf/ocrmypdfNanonets-OCRnanonets/nanonets-ocr
TypeOpen sourceOpen source
Price1 credit / page2 credits / page
Per 1,000 pages$1.00$2.00
Typical speed1.0 s / page1.8 s / page
Best forSearchable PDFs from scansMarkdown with semantic tags
InputsPDF, imagesPDF, images
Languages100+ languages (via Tesseract)Multilingual
Max pages per request1,000500
TablesNo✓ Yes
HandwritingNoNo
MathNo✓ Yes
Bounding boxesNoNo
ConfidenceNoNo
FormsNo✓ Yes
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 OCRmyPDF if

  • You need a validated searchable PDF/A for archiving, with the OCR layer inserted losslessly where possible.
  • You have no GPU. It runs on CPU only and spreads work across all CPU cores.
  • You need 100+ languages through Tesseract language packs and a clear MPL-2.0 licence.

Choose Nanonets-OCR if

  • You need Markdown, HTML, LaTeX or Mermaid output instead of a PDF with a text layer.
  • Your documents have signatures, watermarks or checkboxes, which it tags or converts to unicode symbols.
  • You have flowcharts or org charts and want them as Mermaid code, or equations with inline and display LaTeX kept apart.

Key differences

OCRmyPDFNanonets-OCR
What it doesAdds a Tesseract OCR text layer to scanned PDFs and outputs searchable PDF/AImage-to-Markdown model fine-tuned from Qwen2.5-VL-3B
Output formatsSearchable PDF, plain textMarkdown, HTML, LaTeX, Mermaid
LicenceMPL-2.0Not stated on the Nanonets-OCR2-3B model card (no licence tag); base model Qwen2.5-VL-3B
HardwareCPU only, uses all coresSelf-hosted 3B model (about 4B params on Hugging Face); GPU use not stated in the published material, but one handwriting test ran at 69.1 s per image on an L4
BenchmarksNot publishedolmOCR-bench overall 69.5, DocVQA 89.43, ChartQA 78.56 (OCR2-3B, reported by the authors)
HandwritingListed as not suited to handwritingNot trained on handwriting and can hallucinate; one test saw generation loops with CER up to 2100%
Known failureFillable form fields are removed with --skip-text and default PDF/A output (issue #1790); a hang occurs on PDFs whose forms share one Resources dictionary (issue #1763)Trained on a fixed prompt, so hallucination with custom prompts is expected; loses to Gemini 2.5 Flash in 52.43% of head-to-head Markdown comparisons

Sources

What each one is

OCRmyPDF

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

OCRmyPDF details

Nanonets-OCR

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

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

Questions

More comparisons

OCRmyPDF alternativesNanonets-OCR alternativesBest OCR models in 2026