OCRRouter

Kraken vs Chunkr

Pick Kraken for historical manuscripts, early prints and right-to-left scripts such as Arabic and Hebrew, where you can train layout and recognition on your own material. Pick Chunkr for parsing PDFs, slides and Word files into Markdown, HTML or JSON chunks for RAG. The main split is the job: Kraken transcribes hard historical material, Chunkr turns modern documents into structured output.

Krakenkraken/krakenChunkrchunkr/chunkr
TypeOpen sourceOCR API
Price1 credit / page14 credits / page
Per 1,000 pages$1.00$14.00
Typical speed1.5 s / page3.0 s / page
Best forHistorical documents and manuscriptsLayout-aware chunks for RAG
InputsimagesPDF, images, DOCX
LanguagesMany scripts, incl. right-to-leftMultilingual
Max pages per request5001,000
TablesNo✓ Yes
Handwriting✓ YesNo
MathNo✓ Yes
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 Kraken if

  • Your material is historical manuscripts or early prints, or non-Latin scripts with right-to-left, BiDi or top-to-bottom support.
  • You need ALTO, PageXML, abbyyXML or hOCR output for digital humanities tools.
  • You want to train layout analysis, reading order and character recognition yourself, and can start from community models on Zenodo.

Choose Chunkr if

  • You are chunking PDFs, PPT, Word files or images for RAG and want Markdown, HTML or JSON with bounding boxes.
  • You want a hosted cloud API, or a self-hosted version you run with Docker on GPU, CPU-only or Mac ARM.
  • You run on Windows-based or mixed teams' infrastructure via Docker, or already have an OpenAI-compatible LLM endpoint to plug in.

Key differences

KrakenChunkr
Best forHistorical manuscripts, early prints, Arabic, Hebrew and other RTL scriptsRAG chunking of PDFs and slides, self-hosted layout parsing
LicenceApache-2.0AGPL-3.0 or commercial licence; closed-source use needs the commercial one
How you run itPython library installed with pip, run locallyCloud API, or self-host via Docker with an external OpenAI-compatible LLM endpoint
PriceFree, self-hosted (you pay for compute)Hosted credit-based plans, prices not verified; open-source version free (you pay for compute and an LLM endpoint)
Output formatsALTO, PageXML, hOCR, plain text, bounding boxesHTML, Markdown, JSON, bounding boxes
Platform and inputLinux or macOS, x64 or ARM, Python 3.10 to 3.13; no Windows supportAccepts PDF, PPT, Word and images; Excel not supported in the open-source version
Known issuesInstall dependency conflict (issue #816); learned line order lost in segmentation and text output (issue #820); usually needs training for good accuracyOpen-source version uses community models, less accurate than the cloud's in-house models

Sources

What each one is

Kraken

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

Kraken details

Chunkr

Document parsing API that segments and chunks pages for RAG. Self-hostable version available.

Chunkr 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": "kraken/kraken","fallbacks": ["chunkr/chunkr"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

Kraken alternativesChunkr alternativesBest OCR models in 2026