I understand you'd like to change the LLM, STT, or TTS model configurations during an active conversation. While it's not possible to directly update these settings on the current agent instance mid-conversation, there’s a great way to achieve what you need by transferring control to a new agent instance with the models you want. Here are two friendly methods you can use: • Using `llm.handoff()`: This lets you return a new agent instance from a tool call, smoothly handing over the conversation to that new agent. • Using `session.update_agent()`: This approach replaces the current agent in your session with a new one, and it will automatically trigger the

on_exit()

method of the old agent and the

on_enter()

method of the new one. If you'd like, I’m here to help with more details or examples on how to set these up. Just let me know! Sources: agents/src/voice/agent_activity.ts | multi-agent/long_or_short_agent.py | livekit-plugins/livekit-plugins-openai/livekit/plugins/openai/realtime/real...