Mathpix Convert vs Florence-2
Pick Mathpix Convert for math-heavy papers and textbooks, where you need whole PDFs turned into LaTeX, Markdown or DOCX, including handwritten equations. Pick Florence-2 for English scene text and short labels with bounding boxes, run on your own hardware under an MIT licence. The main split is output: Convert returns document structure, Florence-2 returns only text and boxes.
Mathpix Convertmathpix/convert | Florence-2microsoft/florence-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 | 0.8 s / page |
| Best for | Math and scientific papers | Lightweight OCR with boxes |
| Inputs | PDF, images | images |
| Languages | 30+ languages | English |
| Max pages per request | 500 | 500 |
| Tables | ✓ Yes | No |
| Handwriting | ✓ Yes | No |
| Math | ✓ Yes | No |
| 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 LaTeX, Mathpix Markdown, DOCX or HTML from math, tables and chemistry, not just plain text.
- You are converting whole PDFs or textbooks and want direct DOCX or LaTeX export.
- You have handwritten equations, which Convert lists as a good fit and Florence-2 lists as a poor one.
Choose Florence-2 if
- You want to self-host under an MIT licence and pay only for compute, with no API key fee.
- You need quadrilateral boxes per text line from OCR_WITH_REGION, or one small model (0.23B base, 0.77B large) that also does captioning and detection.
- Your text is English scene text or short labels and you want lightweight edge OCR.
Key differences
| Mathpix Convert | Florence-2 | |
|---|---|---|
| How you run it | Hosted API | Self-hosted via Transformers |
| Licence | Proprietary | MIT |
| 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 and bounding boxes |
| Document structure | Math, tables and chemistry to LaTeX and Markdown | No Markdown, tables or LaTeX |
| Handwriting | Handwritten equations listed as a good fit | Spelling errors and missing spaces on cursive handwriting in an HF blog test |
| Languages and benchmarks | No published accuracy benchmark on general OCR leaderboards | Chinese output mostly wrong because the vocabulary lacks CJK characters; TextCaps CIDEr 72.8 zero-shot (large, reported by the authors) |
Sources
- Mathpix billing docs
- Mathpix OCR API
- Florence-2-large model card
- HF blog OCR test
- HF discussion Chinese very bad
What each one is
Mathpix Convert
Specialist for math, science papers and handwritten formulas. Outputs LaTeX.
Mathpix Convert detailsFlorence-2
Small Microsoft vision model (0.23B and 0.77B) with OCR and OCR-with-regions tasks that return text with boxes.
Florence-2 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": ["microsoft/florence-2"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Mathpix Convert alternativesFlorence-2 alternativesBest OCR models in 2026