Skip to main content
localai.io
Chapter 1

Start here

Note
Needs a build newer than 4.12.1. In the current development builds.

This section takes you from a fresh install to a working chat, a working API call and your first agent. Follow the pages in order the first time.

Tip

On Linux and macOS, install LocalAI with the script curl -sSL https://localai.io/install.sh | sh. Use Docker on Windows or when you prefer a container. See the install guide.

The path

  1. Quickstart: start LocalAI, install a model and get a first chat reply in the web interface and from the API.
  2. Install: every way to install LocalAI, one page per platform.
  3. Set up models: install models from the gallery, Hugging Face, OCI images or local files.
  4. Try it out: copy-paste examples for the OpenAI-compatible endpoints.
  5. Customize a model: edit the model configuration and the prompt template.
  6. Build your first agent: turn a model into a saved agent that a chat, a schedule or a bot can use.

If something fails, see Troubleshooting.