OCRRouter

Chandra vs MonkeyOCR

Pick Chandra for multilingual scans, forms and PDF-to-Markdown for RAG, and for anything commercial that you can run under its licence or through the hosted API. Pick MonkeyOCR for English and Simplified Chinese PDFs, especially academic papers with formulas and tables, when the work is research or evaluation. The main split is licence and language: MonkeyOCR v1 weights are non-commercial only and are not fully supported for handwriting, photos or other languages.

Chandradatalab/chandraMonkeyOCRmonkeyocr/monkeyocr
TypeOpen sourceOpen source
Price2 credits / page2 credits / page
Per 1,000 pages$2.00$2.00
Typical speed2.0 s / page1.5 s / page
Best forComplex layouts and handwritingChinese and English PDFs
InputsPDF, imagesPDF, images
Languages40+ languagesChinese, English
Max pages per request500500
Tables✓ Yes✓ Yes
Handwriting✓ YesNo
Math✓ Yes✓ Yes
Bounding boxes✓ Yes✓ Yes
ConfidenceNoNo
Forms✓ YesNo
Multilingual✓ YesNo
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 Chandra if

  • Chandra 2 averages 77.8% across 43 languages, so it fits multilingual scanned documents where MonkeyOCR is not fully supported.
  • You can ship it commercially: the weights are free for companies under $2M funding or revenue, and the hosted Datalab API costs $1, $3 or $7.50 per 1,000 pages.
  • It returns Markdown, HTML and JSON with bounding boxes, which suits PDF to Markdown for RAG pipelines and documents with tables, forms and math.

Choose MonkeyOCR if

  • Pro-3B reports the best overall OmniDocBench edit distance among the models it was compared with, in English (0.138) and Chinese (0.206), so it suits English and Simplified Chinese PDFs.
  • 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.
  • It outputs LaTeX as well as Markdown, JSON and HTML, which helps with academic papers that have formulas and tables.

Key differences

ChandraMonkeyOCR
LicenceCode Apache-2.0; weights modified OpenRAIL-M, free for research, personal use and companies under $2M funding or revenueCode Apache-2.0; v1 weights research and non-commercial only, commercial use needs a written licence
How you run itSelf-host with vLLM or Hugging Face, or use the hosted Datalab APISelf-host only, via LMDeploy or Docker
PriceSelf-hosted: free within licence terms. API: $1, $3 or $7.50 per 1,000 pages; free tier gives $20/monthFree for research when self-hosted; commercial use needs a separate licence from the authors
HardwareSingle NVIDIA GPU via vLLM; benchmarked on one H100 80GB, with presets for A100, L40S and RTX 4090/5090RTX 3090/4090, A6000, H800, A100; RTX 4060 8GB with quantized models
BenchmarkolmOCR-bench overall 85.8 (Chandra 2); 77.8% average on 43 languagesOmniDocBench edit distance 0.138 EN / 0.206 ZH; table TEDS 81.5 EN / 87.5 ZH (pro-3B)
Languages and inputStrong multilingual coverage, but weak on Urdu (43.2%) and Sanskrit (51.1%)Not fully supported: photographed text, handwriting, Traditional Chinese and multilingual text
Known failureToken-loop hallucinations reported on 5-6 of about 60 Russian/Kazakh scans; Old Scans only 51.1 on olmOCR-benchPro-3B runs under 1 page/s (about 0.49 pages/s on a 3090); Docker on 20/30/40-series and V100 needs a patched image

Sources

What each one is

Chandra

Datalab's OCR model that keeps layout and handles forms, tables and handwriting.

Chandra details

MonkeyOCR

Open-source document parser that splits work into structure, recognition and relation steps.

MonkeyOCR 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": "datalab/chandra","fallbacks": ["monkeyocr/monkeyocr"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

Chandra alternativesMonkeyOCR alternativesBest OCR models in 2026