# MedRun

> MedRun is a library of medical-imaging interpretation models: upload a study, get structured findings, measurements and DICOM SR/SEG. Use it from the console or the API.

Vocabulary: a **model** is an exam type (`angio/coronary`); an **endpoint** is a separate run (`interpret`, or a tool such as `qca`); one run returns one result organised in **sections**, at one price per request.

## Models

- [Coronary Angiography](https://medrun.ai/models/angio/coronary/llms.txt): Coronary angiography read: vessels, SYNTAX segments and stenosis with QCA evidence.

## Docs

- [Introduction](https://medrun.ai/docs): What MedRun is and how requests, inputs and results fit together.
- [Quickstart](https://medrun.ai/docs/quickstart): Run your first model from the console or with one API call.
- [Authentication](https://medrun.ai/docs/api/authentication): API keys, scopes and the Authorization header.
- [Queue API](https://medrun.ai/docs/api/queue): Submit a request, follow its status and fetch the result.
- [Uploading files](https://medrun.ai/docs/api/files): Presigned and multipart uploads for DICOM folders, CD exports and zips.
- [Webhooks](https://medrun.ai/docs/api/webhooks): Signed callbacks when a request completes, fails or flags a critical finding.
- [Result format](https://medrun.ai/docs/result-envelope): The result envelope: sections, findings, measurements, guidance, report and artifacts.
- [MCP server](https://medrun.ai/docs/mcp): Use MedRun models from Claude, ChatGPT and coding tools.

## Optional

- [Explore all models](https://medrun.ai/explore)
- [Pricing](https://medrun.ai/pricing)