Chandra vs OCR.space
Pick Chandra for multilingual scanned documents and PDF to Markdown for RAG, where layout, tables, forms and math matter. Pick OCR.space for low-volume jobs, quick prototypes and making scanned PDFs searchable through a hosted API with a free tier. The main split is that Chandra is an open-weights model you can self-host on your own GPU, while OCR.space is a proprietary hosted API.
Chandradatalab/chandra | OCR.spaceocrspace/ocr-space | |
|---|---|---|
| Type | Open source | OCR API |
| Price | 2 credits / page | 1 credit / page |
| Per 1,000 pages | $2.00 | $1.00 |
| Typical speed | 2.0 s / page | 1.5 s / page |
| Best for | Complex layouts and handwriting | Cheap simple OCR |
| Inputs | PDF, images | PDF, images |
| Languages | 40+ languages | 20+ languages |
| Max pages per request | 500 | 100 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | ✓ Yes | No |
| Math | ✓ Yes | No |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | No | No |
| Forms | ✓ Yes | No |
| Multilingual | ✓ Yes | ✓ Yes |
| On OCRRouter | Available | Available |
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
- You want to self-host on a single NVIDIA GPU via vLLM, with presets for A100, L40S and RTX 4090/5090.
- You need structured output: Markdown, HTML or JSON with layout chunks, bounding boxes and labels per block.
- You process multilingual scans, where Chandra 2 averages 77.8% across 43 languages on the authors' benchmark.
Choose OCR.space if
- You want a free tier of 25,000 requests per month with no GPU to run.
- You need searchable PDFs or per-word overlay coordinates from a hosted API.
- You want a paid plan with a published uptime guarantee: PRO plans include a 100% uptime or money back SLA.
Key differences
| Chandra | OCR.space | |
|---|---|---|
| How you run it | Self-host the open weights (vLLM or Hugging Face) or use the hosted Datalab API | Hosted API only; an on-premise version is sold separately |
| Licence | Code Apache-2.0; weights modified OpenRAIL-M, free for research, personal use and companies under $2M funding/revenue | Proprietary |
| Price | Self-hosted: free within licence terms, you pay for compute. Datalab API: $1, $3 or $7.50 per 1,000 pages depending on mode | Free: 25,000 requests/month. PRO $30/month for 300,000 requests. PRO PDF $60/month. Enterprise from $999/month |
| Limits | Not published as per-file limits; throughput is about 1.44 pages/s on one H100 with vLLM | Free tier: 1 MB files and 3 PDF pages. PRO: 5 MB. PRO PDF: 100 MB+ and 999+ pages |
| Output formats | Markdown, HTML, JSON, bounding boxes | JSON, plain text, bounding boxes, searchable PDF, Markdown |
| Benchmark | olmOCR-bench overall 85.8 for Chandra 2 (authors' figure) | Not published |
| Known failure | Token-loop hallucinations reported on some Russian/Kazakh scans; weak on Urdu (43.2%), Sanskrit (51.1%) and old degraded scans | Free searchable PDFs carry a watermark; Engine 3 is slower on large files with less precise coordinates; Engine 2 does not support Greek |
Sources
What each one is
Chandra
Datalab's OCR model that keeps layout and handles forms, tables and handwriting.
Chandra detailsOCR.space
Low-cost OCR API with a large free tier. Returns text, word boxes and searchable PDFs.
OCR.space detailsTry both on your own file
Run the file once with each model and compare the output.
Use one, fall back to the other
{"model": "datalab/chandra","fallbacks": ["ocrspace/ocr-space"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Chandra alternativesOCR.space alternativesBest OCR models in 2026