OCR API for STEM

Image scanning software for STEM companies. Make creating and searching STEM content easy for your end users.

Example API request
Example API request

Battle-tested with deep functionality, including math equations, chemical diagrams, tables, and full PDF document conversion.

Compatible with industry standards like PDFs, LaTeX, Asciimath, MathML, Markdown, HTML, and SMILES (for chemistry).

The most trusted, innovative, and limit-pushing OCR provider for STEM

latency

Low latency

We are constantly pushing the boundaries of what is possible for image recognition while maintaining extremely low latencies.

Medal

Industry leader

Due to our exceptional accuracy and constant innovation, we are the most trusted brand for OCR among leaders in online math education.

graph

Battle-tested reliability

We process over 10 million images every day. Our system is battle tested and reliable with 99.9%+ uptime.

Global availability enables low latencies

We have 3 AWS server regions to ensure the fastest possible processing speeds for our customers in North/South America (us-east-1), Europe (eu-central-1), and Asia (ap-southeast-1). Requests are routed to the nearest region by default.

Map of our AWS server locations
Map of our AWS server locations

All Mathpix OCR API endpoints

  • v3/text

    Process an image

    This is our most widely used endpoint. You can use this endpoint to OCR individual images of handwritten and printed equations, handwritten and printed text, tables, and diagrams to get digital formats like LaTeX, Asciimath, and MathML.

    You can also request additional information about the data in the image like which alphabets were detected (we support most foreign languages), line data, word data, and confidence.

    This is also the endpoint you want to process images of diagrams like geometry and chemistry. At this time we only support triangles for geometry, which are represented via vertices, edges, and labels.

    Developer Docs
  • Process an image using the v3/text API endpoint
  • v3/strokes

    Process strokes (digital ink)

    Use this endpoint to add stroke recognition (ie. digital ink) to your app. This endpoint works for all math, and for Hindi and Latin alphabet language text. You can process the data for each individual stroke, or you can use a session_id for live drawing capabilities in your app (this is also the more cost-effective option).

    Developer Docs
  • v3/pdf

    Process a PDF

    Use this endpoint to OCR PDFs and convert them to Markdown, LaTeX, and DOCX. This endpoint works asynchronously, since large PDF files can take several minutes to process. This endpoint accepts PDF files and URLs in the request.

    Since requests are asynchronous, you can check the processing status of a PDF using the PDF ID returned by the API. Once the PDF has been processed, you can convert it to your desired format. The PDF endpoint can also return line-by-line data, which can be useful for building novel experiences on top of original PDFs. A separate PDF results endpoint ( v3/pdf-results) can be used to query information about a certain PDF, or all your PDFs.

    Developer Docs
  • Process an image using the v3/text API endpoint
  • v3/batch

    Process a batch of images

    Use this endpoint to process a multiple images in a single POST request. The request will return a batch ID which be queried with a GET request once an appropriate amount of time has passed for the images to process. You should only use the batch endpoint if your workflow is not latency sensitive (for example, you are digitizing a textbook and need to process all images of equations). Please also note that batch requests are priced per image, not per request.

    Developer Docs
  • Process a batch of images using the v3/batch endpoint

EdTech industry leaders trust Mathpix

Gradescope
Mathway
Chegg
Byjus
Coursehero
Equatio
Facebook
Ruangguru
Symbolab
Toppr
Wjtb
Benetech
FEATURES

Digitize entire PDFs

PDFs are first converted into Mathpix Markdown, from which they can be exported to DOCX / MS Word, LaTeX, and PDF. Text, diagrams, equations, and tables, are extracted from the PDF.

Original PDF

Example PDF
271218 Created with Sketch.

Extracted Mathpix Markdown

Example equation
271218 Created with Sketch.

Rendered result

Example equation

Supported image types


Use Mathpix OCR to very accurately convert images of simple and complicated printed and handwritten math, text, tables, and chemical diagrams.


We can also recognize all the world's most spoken languages like English, Vietnamese, Spanish, French, German, Hindi, Chinese, Japanese, Russian, Korean, Thai, and more.

Printed math

Example equation

Handwritten math

Example equation

Word problems

Example equation

Chemical equation

Example equation

Multiple choice questions

Example equation

Tables

Example image

Printed Foreign Alphabet text

Example imageExample imageExample image

Chemical Diagram

Example image

Printed English

Example image

You can find more example requests and types of images that you can process with Mathpix OCR in the User Guide.

Use cases for Mathpix OCR

Solving and Search Apps

With a long tail of advanced math, handwriting, and foreign language features, Mathpix OCR is the #1 choice for solving, tutoring, and search apps like Mathway, Doubtnut, and Toppr. These apps help students of all levels around the world learn math, find tutors, and get the help that they need.

Grading and Assessment Platforms

Whether they provide immediate answers or grade questions in bulk, grading and assessment platforms like Gradescope use Mathpix OCR in their platforms to make it faster and easier for professors to distribute and grade homework and exams with AI.

Publishing, Accessibility and LMS

Publishing companies use Mathpix OCR to create digital STEM educational material like online math textbooks. Accessible education companies like Benetech use Mathpix to make math and science educational materials accessible to all students.

Bulk digitization (on-prem available)

Our APIs are useful for bulk digitization of images or PDFs. We also offer an on-premise solution for bulk PDF processing for companies like London Stock Exchange, who need to securely process documents in their private cloud.

Ready to get started?

Choose the plan that’s right for you and your team.

Monthly
Pay As You Go

$19.99 one-time setup fee
1000 free API calls per month for testing
Visualization Dashboard
Pay only for what you use
Billing on 1st of each month for the previous month’s API usage

Enterprise

Long-term contracts
24/7 support
Custom recognition features
SLAs
On prem (PDF and document conversion processing)
Discounted prices for high volume

Yearly License
On-Prem

Fully contained AMI
Unlimited usage
No access to outside internet needed
Maintenance included
Software improvements included
GCP or Azure available upon request
Docker image available upon request

API pricing by route

We price differently based on which endpoints are used. Usage tiers serve as a volume discount.

Process image (v3/text)

Process strokes (without live updates)(v3/strokes)

Process equation image (v3/latex)

Process batch (v3/batch)*

Get results (v3/ocr-results)

$0/request (0-1K requests)
$0.004/request (1-100K requests)
$0.002/request (100-300K requests)
$0.001/request (300K+ requests)

*Note: each item in a batch request counts as one request

Process PDF (v3/pdf)

$0.1/page (0-1K pages)
$0.05/page (1K+ pages)

Digital ink with live updates included (v3/strokes using stroke session ID)

$0/session (0-1K sessions)
$0.01/session (1-100K sessions)
$0.008/session (100K-1000K sessions)
$0.005/session (1000K+ sessions)