Kraken vs MonkeyOCR
Pick MonkeyOCR for English and Simplified Chinese PDFs such as academic papers with formulas and tables, where it outputs Markdown, JSON and LaTeX. Pick Kraken for historical manuscripts, early prints and right-to-left scripts, where you can train layout and recognition on your own material. Check the licence first: MonkeyOCR v1 weights are non-commercial without a written license, while Kraken is Apache-2.0.
Krakenkraken/kraken | MonkeyOCRmonkeyocr/monkeyocr | |
|---|---|---|
| Type | Open source | Open source |
| Price | 1 credit / page | 2 credits / page |
| Per 1,000 pages | $1.00 | $2.00 |
| Typical speed | 1.5 s / page | 1.5 s / page |
| Best for | Historical documents and manuscripts | Chinese and English PDFs |
| Inputs | images | PDF, images |
| Languages | Many scripts, incl. right-to-left | Chinese, English |
| Max pages per request | 500 | 500 |
| Tables | No | ✓ Yes |
| Handwriting | ✓ Yes | No |
| Math | No | ✓ Yes |
| Bounding boxes | ✓ Yes | ✓ 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 Kraken if
- You work with historical manuscripts, early prints, or non-Latin scripts such as Arabic and Hebrew, including right-to-left, BiDi and top-to-bottom text.
- You want to train layout analysis, reading order and character recognition on your own material.
- You need ALTO, PageXML or hOCR output for digital humanities tools, and an Apache-2.0 licence with no commercial restriction.
Choose MonkeyOCR if
- You process English or Simplified Chinese PDFs, especially academic papers with formulas and tables.
- You want Markdown, JSON with block coordinates, LaTeX or HTML output without training a model first.
- You have a single GPU: it runs on an 8GB RTX 4060 with quantized weights, and the 1.2B model is about 36% faster than pro-3B for about 1.6% lower accuracy.
Key differences
| Kraken | MonkeyOCR | |
|---|---|---|
| Licence | Apache-2.0 | Code Apache-2.0; v1 weights research and non-commercial only, commercial use needs a written license |
| How you run it | Python library installed with pip on Linux or macOS (x64 or ARM, Python 3.10 to 3.13). No Windows support. | Self-hosted through LMDeploy or Docker on a GPU. Docker on 20/30/40-series and V100 needs a patched image to avoid a shared-memory error. |
| Best-fit documents | Historical manuscripts, early prints, Arabic, Hebrew and other RTL scripts | English and Simplified Chinese PDFs, academic papers with formulas and tables |
| Output formats | ALTO, PageXML, hOCR, plain text, bounding boxes | Markdown, JSON, LaTeX, HTML, bounding boxes |
| Benchmark | Not published | OmniDocBench (pro-3B, author-reported): overall edit distance 0.138 EN / 0.206 ZH (lower is better), table TEDS 81.5 EN / 87.5 ZH |
| Training needed | Usually needs training on your own material for good accuracy | Not required, but pro-3B throughput is under 1 page/s (about 0.49 pages/s on a 3090) |
| Known failure | Install can fail with a dependency conflict (issue #816). Learned line order is lost in segmentation and text output (issue #820). | Not fully supported: photographed text, handwriting, Traditional Chinese and multilingual text. |
Sources
What each one is
Kraken
OCR engine for historical documents and manuscripts, with trainable models for unusual scripts.
Kraken 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": "kraken/kraken","fallbacks": ["monkeyocr/monkeyocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Kraken alternativesMonkeyOCR alternativesBest OCR models in 2026