Mathpix Convert vs Chandra
Pick Mathpix Convert for math-heavy papers and textbooks where you need LaTeX or DOCX out of whole PDFs, and for handwritten equations. Pick Chandra for multilingual scans and PDF to Markdown pipelines, especially if you want to self-host or pay per 1,000 pages. The main split is subject matter and deployment: Mathpix is a hosted STEM API, Chandra is an open-weights layout model you can run yourself.
Mathpix Convertmathpix/convert | Chandradatalab/chandra | |
|---|---|---|
| Type | OCR API | Open source |
| Price | 7 credits / page | 2 credits / page |
| Per 1,000 pages | $7.00 | $2.00 |
| Typical speed | 1.5 s / page | 2.0 s / page |
| Best for | Math and scientific papers | Complex layouts and handwriting |
| Inputs | PDF, images | PDF, images |
| Languages | 30+ languages | 40+ languages |
| Max pages per request | 500 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | ✓ Yes | ✓ Yes |
| Math | ✓ Yes | ✓ Yes |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | ✓ Yes | No |
| Forms | No | ✓ Yes |
| Multilingual | No | ✓ 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 Mathpix Convert if
- It converts math, tables and chemistry into LaTeX and Markdown, and exports DOCX and LaTeX directly from whole PDFs.
- It handles handwritten equations, which suits textbooks and student work.
- Pricing is simple to predict: PDF pages cost $0.005 each (0-1M) and $0.0035 above 1M, and images cost $0.002 each.
Choose Chandra if
- You can self-host the open weights with vLLM or Hugging Face on a single NVIDIA GPU, or use the hosted Datalab API.
- Hosted cost is $1, $3 or $7.50 per 1,000 pages depending on mode, with a free tier of $20 per month.
- It scores 85.8 on olmOCR-bench and 77.8% on a 43-language multilingual benchmark (authors' figures), and returns bounding boxes and labels per block.
Key differences
| Mathpix Convert | Chandra | |
|---|---|---|
| How you run it | Hosted API only | Self-host open weights (vLLM or Hugging Face) or hosted Datalab API |
| Licence | Proprietary | Code Apache-2.0; weights modified OpenRAIL-M, free for research, personal use and companies under $2M funding/revenue |
| Price | PDF $0.005 per page (0-1M), $0.0035 above 1M; images $0.002 each; one-time $19.99 API key fee | Hosted $1, $3 or $7.50 per 1,000 pages; self-hosted free within licence terms, you pay for compute |
| Output formats | LaTeX, Markdown, DOCX, HTML, JSON | Markdown, HTML, JSON, bounding boxes |
| Benchmarks | Not published | olmOCR-bench 85.8; 43-language average 77.8%; 90-language average 72.7% (authors' figures) |
| Focus | STEM: math, tables, chemistry, handwritten equations | Multilingual scans, tables, forms and math, PDF to Markdown for RAG |
| Known failure | Images with more than 12 text rows may be billed at the PDF page rate | Token-loop hallucinations reported on some Russian/Kazakh scans; weak on Urdu (43.2%), Sanskrit (51.1%) and old scans |
Sources
What each one is
Mathpix Convert
Specialist for math, science papers and handwritten formulas. Outputs LaTeX.
Mathpix Convert detailsChandra
Datalab's OCR model that keeps layout and handles forms, tables and handwriting.
Chandra 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": "mathpix/convert","fallbacks": ["datalab/chandra"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Mathpix Convert alternativesChandra alternativesBest OCR models in 2026