Mathpix Convert vs GOT-OCR 2.0
Pick Mathpix Convert when you need whole PDFs or images of math-heavy papers and textbooks turned into LaTeX, Markdown or DOCX through a hosted API, including handwritten equations. Pick GOT-OCR 2.0 when you want a free, self-hosted model for formulas, sheet music, chemical structures or region-specific OCR and can run CUDA hardware. The main split is paid hosted API versus self-hosted model where you pay only for compute.
Mathpix Convertmathpix/convert | GOT-OCR 2.0stepfun/got-ocr-2 | |
|---|---|---|
| Type | OCR API | Open source |
| Price | 7 credits / page | 1 credit / page |
| Per 1,000 pages | $7.00 | $1.00 |
| Typical speed | 1.5 s / page | 1.5 s / page |
| Best for | Math and scientific papers | Formatted text, formulas and charts |
| Inputs | PDF, images | images |
| Languages | 30+ languages | Chinese, English |
| Max pages per request | 500 | 100 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | ✓ Yes | No |
| Math | ✓ Yes | ✓ Yes |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | ✓ Yes | No |
| Forms | No | No |
| Multilingual | No | No |
| 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
- You need DOCX or LaTeX export straight from whole PDFs, with HTML and JSON also available.
- Your documents are math-heavy papers and textbooks, or contain handwritten equations, tables and chemistry.
- You want a hosted API with no GPU to manage, at $0.005 per PDF page up to 1M pages and $0.0035 above that.
Choose GOT-OCR 2.0 if
- You want no per-page fee: the model is free to self-host and you pay only for compute.
- You need region OCR guided by box coordinates or color, or outputs such as TikZ, SMILES and kern for geometry, chemical structures and sheet music.
- You want to run a 580M-parameter model you control, available in Hugging Face Transformers with batched inference.
Key differences
| Mathpix Convert | GOT-OCR 2.0 | |
|---|---|---|
| How you run it | Hosted API | Self-hosted; README lists CUDA 11.8 and PyTorch 2.0.1, with a llama.cpp port for CPU users |
| Licence | Proprietary | Code Apache-2.0; training data CC BY-NC 4.0 |
| Price | PDF $0.005 per page (0-1M), $0.0035 above 1M; images $0.002 each; one-time $19.99 API key activation fee | Free; you pay for compute |
| Output formats | LaTeX, Markdown, DOCX, HTML, JSON | Plain text, Markdown, LaTeX, TikZ, SMILES, kern |
| Content covered | STEM content: math, tables and chemistry; handwritten equations | Text, formulas, tables, charts, sheet music and geometry; English and Chinese documents |
| Known failure modes | Images with more than 12 text rows may be billed at the PDF page rate; billed monthly in arrears | Output can loop characters midway (issue #226); returns wrong text on blank images (issue #235); format mode is much worse than plain OCR mode (issue #317) |
| Published benchmarks | Not published on general OCR leaderboards | Not published in the README |
Sources
What each one is
Mathpix Convert
Specialist for math, science papers and handwritten formulas. Outputs LaTeX.
Mathpix Convert detailsGOT-OCR 2.0
Unified OCR model that reads plain text, tables, formulas, sheet music and charts.
GOT-OCR 2.0 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": ["stepfun/got-ocr-2"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Mathpix Convert alternativesGOT-OCR 2.0 alternativesBest OCR models in 2026