Direct LLM Chat

Chat with your local LLM models directly from the dashboard. Streaming responses, session history, and configurable model routing.

Chat capabilities

Chat uses the 'liaison' capability routing to connect to your configured Ollama models.

Connected MCP servers provide tools the LLM can call: file search, code search, project queries, and more.

Responses stream in real-time from local models. No cloud latency.

Chat interactions are logged with full prompt/response pairs and token usage.

Tools the LLM can call during chat: project state, code search, git, and more.

Chat uses the liaison capability for model selection.

Chat interactions are logged for audit.