Welcome
How to install, run, and operate your private AI infrastructure with OwnLLM.
OwnLLM turns one GPU machine into a private multi-user AI platform: an OpenAI-compatible API, SSO, audit logs, and zero-config networking.
These docs cover three surfaces:
- CLI —
ownllmheadless agent for Linux servers, VPS, and power users. - Desktop app — Atlas, the native Tauri client for Mac, Windows, and Linux desktops.
- Public API — the OpenAI-compatible HTTP endpoint your team's tools talk to.
Where to start
- New here? Read the Getting Started overview.
- Installing on a server? Jump to the CLI section once it lands.
- Connecting Cursor or Claude Code? See the API integrations once they land.
Need help?
Email hello@ownllm.app or open an issue at github.com/impulse-studio/ownllm.