OCRmyPDF vs MonkeyOCR
Pick OCRmyPDF to make scanned PDFs searchable and keep them as PDF/A, on CPU only, under MPL-2.0. Pick MonkeyOCR to turn English or Simplified Chinese PDFs with tables and formulas into Markdown or JSON, but only for research unless you get a written commercial license. The main split is output: OCRmyPDF adds a text layer to the same PDF, while MonkeyOCR parses layout into structured data.
OCRmyPDFocrmypdf/ocrmypdf | MonkeyOCRmonkeyocr/monkeyocr | |
|---|---|---|
| Type | Open source | Open source |
| Price | 1 credit / page | 2 credits / page |
| Per 1,000 pages | $1.00 | $2.00 |
| Typical speed | 1.0 s / page | 1.5 s / page |
| Best for | Searchable PDFs from scans | Chinese and English PDFs |
| Inputs | PDF, images | PDF, images |
| Languages | 100+ languages (via Tesseract) | Chinese, English |
| Max pages per request | 1,000 | 500 |
| Tables | No | ✓ Yes |
| Handwriting | No | No |
| Math | No | ✓ Yes |
| Bounding boxes | No | ✓ Yes |
| Confidence | No | No |
| Forms | No | No |
| Multilingual | ✓ Yes | No |
| On OCRRouter | Available | Available |
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 searchable PDF/A output for archives, and it writes the OCR layer without re-encoding other content where possible.
- You have no GPU: it runs on CPU only and spreads work across all cores.
- You need 100+ languages through Tesseract language packs, or a licence (MPL-2.0) that does not restrict commercial use.
Choose MonkeyOCR if
- You need Markdown, JSON, LaTeX, HTML or bounding boxes, which OCRmyPDF does not produce.
- Your documents are English or Simplified Chinese papers with formulas and tables, where pro-3B scores 0.138 / 0.206 overall edit distance on OmniDocBench.
- You have a GPU, even an 8GB RTX 4060 with quantized weights, and can pick pro-1.2B for about 36% more speed at about 1.6% lower accuracy.
Key differences
| OCRmyPDF | MonkeyOCR | |
|---|---|---|
| Licence | MPL-2.0 | Code Apache-2.0; v1 weights for research and non-commercial use only, commercial use needs a written license |
| Output | Searchable PDF/A, plain text | Markdown, JSON, LaTeX, HTML, bounding boxes |
| Hardware | CPU only, uses all cores | GPU: tested on RTX 3090/4090, A6000, H800, A100; RTX 4060 8GB with quantized models |
| Layout and tables | No layout, table or Markdown output | Table TEDS 81.5 / 87.5 EN/ZH (pro-3B, reported by the authors) |
| Languages | 100+ via Tesseract language packs | English and Simplified Chinese; Traditional Chinese and multilingual text not fully supported |
| Handwriting | Not suited to handwriting | Not fully supported, along with photographed text |
| Known failure | --skip-text with default PDF/A output removes form fields; hangs on PDFs whose forms share one Resources dictionary | Docker on 20/30/40-series and V100 needs a patched image to avoid an LMDeploy shared-memory error; pro-3B runs at about 0.49 pages/s on a 3090 |
Sources
What each one is
OCRmyPDF
Command-line tool that adds a searchable text layer to scanned PDFs, using Tesseract.
OCRmyPDF detailsMonkeyOCR
Open-source document parser that splits work into structure, recognition and relation steps.
MonkeyOCR detailsTry both on your own file
Run the file once with each model and compare the output.
Use one, fall back to the other
{"model": "ocrmypdf/ocrmypdf","fallbacks": ["monkeyocr/monkeyocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
OCRmyPDF alternativesMonkeyOCR alternativesBest OCR models in 2026