Chat with your local model

Runs on the server's GPU via Ollama · nothing leaves your tailnet