Skip to main content

​
What is ParseRouter?

ParseRouter turns documents — PDFs and images — into clean, structured output through MinerU the state-of-the-art open-source document parsing model, served as a production-ready API. You don’t run GPUs, manage model weights, or stitch together OCR pipelines. You send a file, you get back Markdown and structured blocks. And unlike a typical parsing endpoint, you always know exactly which model processed your document — at a flat, per-page price. No black boxes.

​
Why ParseRouter?

State-of-the-art accuracy

Powered by MinerU, a leading open-source model for tables, formulas, and complex layouts.

Transparent pricing

A flat rate of 1 credit per page. You’re only charged for documents that parse successfully.

Know your model

Every request names the engine and model that ran it. Reproducible, auditable, no surprises.

Built for production

Predictable rate limits, explicit error codes, and a simple bearer-token API you can integrate in minutes.

​
How it works

  1. Get an API key from your dashboard (sk-pr-…).
  2. Upload a document to the parse endpoint as multipart form data.
  3. Receive structured output – Markdown plus per-page blocks (titles, paragraphs, tables, and more).
curl -X POST https://api.parserouter.com/v1/mineru/parse \
  -H "Authorization: Bearer sk-pr-..." \
  -F "file=@invoice.pdf" \
  -F "model=mineru2.5-pro-2605"

​
Next steps

Quickstart

Go from API key to your first parsed document in under five minutes.

API Reference

Full details on the parse endpoint, parameters, and response shape.

Credits & billing

How per-page pricing works and when you’re charged.

Limits

Rate limits, the per-document page cap, and timeout guidance.