Mathpix Files API
High-throughput, asynchronous document processing at scale. Convert PDFs and documents to Markdown, DOCX, LaTeX, and more — submit one file or hundreds of thousands in a single request, and have results delivered straight to your own cloud storage.
Built for large-scale document pipelines
Built for Scale
Submit thousands of documents through a single API — up to 200,000 files in one request. No queueing, retry, or chunking infrastructure to build yourself.
Cloud-Native Integrations
Read from and write results directly to your own S3, Google Cloud Storage, or Azure Blob bucket. Keyless — grant access via IAM role, AD app, or service-account impersonation. No secrets shared.
Lower Cost at Volume
Tiered per-page pricing optimized for batch workloads — you pay less as your volume grows. See pricing →
Self-Serve
Start today with your existing API key. Register a data source once, then submit jobs and track them programmatically — no sales call, no onboarding ticket.
From submission to structured data
1. Connect your storage
Optionally grant Mathpix keyless access to your S3, GCS, or Azure bucket so we can read sources and write results back to storage you control.
2. Submit documents
Send one document via POST /files/v1/uri, or up to 200,000 at once via POST /files/v1/jobs — from public URLs or your connected buckets.
3. Track the job
Poll job status and per-file results, filter for errors, and correlate by your own custom_id. Idempotent submission makes retries safe.
4. Collect results
Download Mathpix Markdown, DOCX, LaTeX, HTML and more — or have them written straight to your bucket with relatively-linked cropped images.
Migrating from SCS?
The Files API is the self-serve successor to our Secure Conversion Service — same processing pipeline and output formats, now driven by your own API calls. Existing SCS customers can self-migrate without a support ticket.
Read the migration guideReady to Get Started?
Start with your existing API key, or talk to us about high-volume pricing.