Configure an LLM

Open Settings → LLMs. For the product story — self-hosted and partner-hosted Private AI as an organizationally aware package — see Private AI models.

1. Prefer Custom LLMs for Private AI (OpenAI-compatible / Ollama — self-hosted or partner-hosted).
2. Optionally add Public LLMs (OpenAI, Anthropic, Perplexity, …) as ranked fallbacks.
3. Connect credentials and endpoint details; Test Connection.
4. Prefer Diode connection modes when available: Perimeter Discover or manual bind — so model traffic stays on your private path.
5. Drag to rank providers; put Private AI first when policy requires it.

Without an enabled LLM, chat cannot answer. Pair models with Tools and MCP so the agent can ground answers in organization data.