OCRRouter

Kraken vs Typhoon OCR

Pick Kraken for historical manuscripts, early prints and right-to-left scripts such as Arabic and Hebrew, where you are willing to train on your own material. Pick Typhoon OCR for modern Thai and English documents such as government forms and financial reports, where you want Markdown, HTML tables and LaTeX out of the box. The main split is the material: Kraken targets old and non-Latin scripts, Typhoon OCR targets Thai-English business documents.

Krakenkraken/krakenTyphoon OCRtyphoon/typhoon-ocr
TypeOpen sourceOpen source
Price1 credit / page2 credits / page
Per 1,000 pages$1.00$2.00
Typical speed1.5 s / page1.8 s / page
Best forHistorical documents and manuscriptsThai documents
InputsimagesPDF, images
LanguagesMany scripts, incl. right-to-leftThai, English
Max pages per request500500
TablesNo✓ Yes
Handwriting✓ YesNo
MathNoNo
Bounding boxes✓ YesNo
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, which Kraken was built for.
  • You need right-to-left, BiDi or top-to-bottom scripts such as Arabic and Hebrew.
  • You want ALTO, PageXML or hOCR output for digital humanities tools, with trainable layout analysis, reading order and recognition.

Choose Typhoon OCR if

  • Your documents are Thai government forms, financial reports or Thai books, or mix Thai and English.
  • You need tables as HTML with merged cells, equations as LaTeX and checkboxes marked.
  • You want a choice of self-hosting the weights or calling the Typhoon API as typhoon-ocr-v2.

Key differences

KrakenTyphoon OCR
Target materialHistorical manuscripts, early prints and non-Latin scriptsThai and English documents such as government forms, financial reports and books
How you run itPython library, installed with pip on Linux or macOS (x64 or ARM, Python 3.10 to 3.13). No Windows support.Self-hosted weights, or the Typhoon API as typhoon-ocr-v2
Licence and priceApache-2.0. Free, self-hosted; you pay for compute.Apache-2.0, subject to OpenTyphoon Terms and Conditions. Free, self-hosted; API pricing is not published on the model card.
Output formatsALTO, PageXML, hOCR, plain text and bounding boxesMarkdown, HTML and LaTeX
BenchmarksNot publishedV2 2B: olmOCR-bench 84.7, OmniDocBench v1.6 90.97, ThaiOCRBench 59.9 (reported by the authors)
Training neededUsually needs training on your own material for good accuracyModel is used as released; no training step described
Known failureInstall dependency conflict (issue #816); learned line order lost in segmentation and text output (issue #820)V2 can loop until the token limit on dense multi-column pages; key-information extraction is weakest at 21.0 on ThaiOCRBench

Sources

What each one is

Kraken

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

Kraken details

Typhoon OCR

Open vision model tuned for Thai and English documents, with Markdown output.

Typhoon 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": "kraken/kraken","fallbacks": ["typhoon/typhoon-ocr"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

Kraken alternativesTyphoon OCR alternativesBest OCR models in 2026