Mathpix OCR API Overview
What is Mathpix OCR API?
Mathpix OCR API is a web API for extracting text from images, digitizing documents, and supporting digital ink in your application. It works especially well for content containing scientific notation like math, chemistry, and physics. You can find example images with API responses here.
What you can build with it
| You want to | Use |
|---|---|
| Recognize one image at a time, synchronously | v3/text |
| Convert a document to Markdown, DOCX, LaTeX, HTML, or PPTX | v3/pdf |
| Convert Mathpix Markdown you already have into another format | v3/converter |
| Recognize digital ink as the user draws | v3/strokes |
| Process large archives — up to 200,000 documents per job | files/v1 |
How this guide is organized
- Get started — what the API is, a Quickstart with runnable requests, and how to create an API key.
- Using the API — the endpoint catalog, best practices, rate and page limits, and worked examples.
- Console guide — the parts of console.mathpix.com you will use day to day: the Playground, usage, exports and results, and members and roles.
- Billing — plans and pricing, invoices and payment, and what happens if an invoice goes unpaid.
- Reference — supported languages, supported math characters, privacy, use cases, and the changelog.
API reference
This guide explains concepts, workflows, and the Console. The complete API reference — every request parameter, response field, and error code — lives at docs.mathpix.com.
There is also an official Python SDK (
pip install mpxpy) that wraps authentication, polling, and result downloads.How can I contact support?
Just send us an email at support@mathpix.com.