OCRRouter

MinerU vs Kraken

Pick MinerU to convert modern PDFs, scans and Office files into Markdown or JSON, especially Chinese and English papers or a mixed corpus for RAG. Pick Kraken for historical manuscripts, early prints and right-to-left scripts, where you can train layout and recognition on your own material. The main split is the target material: MinerU is a ready-made parser for current documents, Kraken is a trainable system for historical and non-Latin text.

MinerUopendatalab/mineruKrakenkraken/kraken
TypeOpen sourceOpen source
Price2 credits / page1 credit / page
Per 1,000 pages$2.00$1.00
Typical speed2.0 s / page1.5 s / page
Best forAcademic PDFsHistorical documents and manuscripts
InputsPDF, imagesimages
Languages80+ languagesMany scripts, incl. right-to-left
Max pages per request500500
Tables✓ YesNo
HandwritingNo✓ Yes
Math✓ YesNo
Bounding boxes✓ Yes✓ Yes
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 MinerU if

  • It accepts PDF, images, DOCX, PPTX, XLSX, EPUB, OFD and HTML in one tool, which suits mixed Office and PDF corpora.
  • It outputs Markdown, HTML, LaTeX, DOCX, EPUB and JSON, so the result drops into RAG and publishing pipelines.
  • The basic ONNX tier runs on CPU with about 2 GB RAM, so you can start batch conversion without a GPU.

Choose Kraken if

  • It is built for historical documents and non-Latin scripts, with right-to-left, BiDi and top-to-bottom support.
  • Layout analysis, reading order and character recognition are all trainable on your own material.
  • It is Apache-2.0 licensed and writes ALTO, PageXML, abbyyXML and hOCR, the formats digital humanities tools use.

Key differences

MinerUKraken
Best materialChinese and English academic papers, mixed Office and PDF corporaHistorical manuscripts, early prints, Arabic, Hebrew and other RTL scripts
LicenceMinerU Open Source License (Apache-2.0 based with extra conditions); MinerU2.5 VLM weights are tagged AGPL-3.0Apache-2.0
How you run itLocal CLI, API or WebUI; CPU-only basic tier, or VLM tiers needing 16 GB RAM and a GPU with 8 GB+ VRAMPython library installed with pip on Linux or macOS (x64 or ARM, Python 3.10 to 3.13); no Windows support
Input formatsPDF, images, DOCX, PPTX, XLSX, EPUB, OFD, HTMLNot published
Output formatsMarkdown, HTML, LaTeX, DOCX, EPUB, JSONALTO, PageXML, hOCR, plain text, bounding boxes
BenchmarksolmOCR-Bench overall 75.2 (MinerU 2.5.4); ParseBench mean 45.9 (MinerU2.5)Not published
Known failuresVLM tiers drop merged header rows in tables such as bank statement account lines; repeated content can produce 300 KB figure blocks; full-width Chinese punctuation is converted to half-widthInstall can fail with a dependency conflict; learned line order is lost in segmentation and text output; usually needs training for good accuracy

Sources

What each one is

MinerU

Open-source PDF to Markdown pipeline popular for academic and Chinese documents.

MinerU details

Kraken

OCR engine for historical documents and manuscripts, with trainable models for unusual scripts.

Kraken 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": "opendatalab/mineru","fallbacks": ["kraken/kraken"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

MinerU alternativesKraken alternativesBest OCR models in 2026