OCRRouter

Unlimited OCR vs Chandra

Pick Unlimited OCR for long reports and books that you want parsed in one request under an MIT licence. Pick Chandra for multilingual scans and for documents with tables, forms and math, where you need Markdown, HTML or JSON. The main split is document length and licence versus language coverage and output formats.

Unlimited OCRbaidu/unlimited-ocrChandradatalab/chandra
TypeOpen sourceOpen source
Price2 credits / page2 credits / page
Per 1,000 pages$2.00$2.00
Typical speed1.2 s / page2.0 s / page
Best forBooks and very long documentsComplex layouts and handwriting
InputsPDF, imagesPDF, images
Languages30+ languages40+ languages
Max pages per request1,000500
Tables✓ Yes✓ Yes
HandwritingNo✓ Yes
Math✓ Yes✓ Yes
Bounding boxes✓ Yes✓ Yes
ConfidenceNoNo
FormsNo✓ Yes
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 Unlimited OCR if

  • It parses 40+ pages in a single forward pass with roughly constant memory, so books and long reports go in as one request.
  • The MIT licence allows commercial use with no copyleft and no revenue cap.
  • It returns block types and bounding boxes alongside text, which suits layout-aware extraction and bulk PDF to text for RAG.

Choose Chandra if

  • It publishes multilingual results: 77.8% average across 43 languages, and 72.7% across 90 languages.
  • It outputs Markdown, HTML and JSON, plus layout chunks with bounding boxes and labels per block.
  • A hosted Datalab API is priced per page ($1, $3 or $7.50 per 1,000 pages), with a free tier of $20 per month, so you can skip running GPUs.

Key differences

Unlimited OCRChandra
LicenceMIT, commercial use allowedCode Apache-2.0. Weights modified OpenRAIL-M: free for research, personal use and companies under $2M funding or revenue. Above that you need a commercial licence.
How you run itSelf-host with vLLM, SGLang or transformers. Also offered hosted in Baidu Cloud's OCR product.Self-host with vLLM or Hugging Face. Or use the hosted Datalab API.
PriceFree self-hosted, you pay for compute. Hosted pricing not published.Free self-hosted within licence terms. Datalab API: $1 (fast), $3 (balanced) or $7.50 (accurate) per 1,000 pages.
Output formatsMarkdown, plain text, bounding boxesMarkdown, HTML, JSON, bounding boxes
HardwareNVIDIA GPU with CUDA 12.9 or 13.0. Version 2 under SGLang runs out of memory on 12 GB GPUs.Single NVIDIA GPU via vLLM. Presets for A100, L40S and RTX 4090/5090. Benchmarks run on one H100 80GB.
BenchmarkParseBench mean 46.17. Unlimited OCR 2: 84.6 on MPDocBench and 77.1 on MarathonDocBench (authors' figures).Chandra 2: 85.8 on olmOCR-bench (authors' figure). No shared benchmark with Unlimited OCR is published.
Known failureText repetition and looping (issue #55), hallucinated text (issue #58), and bbox coordinate loops on GB10 GPUs with SGLang (issue #114). Bold and italic formatting is mostly lost. Context capped at 32,768 tokens.Token-loop hallucinations on 5-6 of about 60 Russian and Kazakh scans, with missing pages (issue #51). Weak on Urdu (43.2%), Sanskrit (51.1%) and old scans (51.1 on olmOCR-bench). One reviewer measured 66 seconds per page on an RTX 4090 with the Hugging Face path.

Sources

What each one is

Unlimited OCR

Open-source model from Baidu that reads hundreds of pages in one pass with flat memory use.

Unlimited OCR details

Chandra

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

Chandra 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": "baidu/unlimited-ocr","fallbacks": ["datalab/chandra"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

Unlimited OCR alternativesChandra alternativesBest OCR models in 2026