# Relay

> Run robot policies on hosted GPUs from Python.

- [Overview](https://docs.relayintelligence.org/index.md): Run π0, π0.5 and GR00T robot policies on hosted GPUs from a few lines of Python.
- [Quickstart](https://docs.relayintelligence.org/quickstart.md): Install relay, point it at a server and get your first action chunk.
- [Observations](https://docs.relayintelligence.org/observations.md): What to send each step, for each model, and the options you can add to any request.
- [Control loops](https://docs.relayintelligence.org/control-loop.md): Run a policy on a real robot, episode after episode.
- [Models](https://docs.relayintelligence.org/models.md): The policies WALLE serves and how they differ.
- [API reference](https://docs.relayintelligence.org/api.md): The PolicyClient class and the environment variables it reads.
- [Troubleshooting](https://docs.relayintelligence.org/troubleshooting.md): What each error means and how to fix it.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.