> ## Documentation Index
> Fetch the complete documentation index at: https://phaseo.app/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# LiveKit

> Use Phaseo models in LiveKit Agents.

LiveKit Agents can call any OpenAI-compatible chat endpoint through its OpenAI plugin. Point that plugin at Phaseo to use Phaseo model ids and routing.

## Python

```bash theme={null}
uv add "livekit-agents[openai]"
```

```python theme={null}
import os
from livekit.agents import AgentSession
from livekit.plugins import openai

session = AgentSession(
    llm=openai.LLM(
        model="openai/gpt-5-nano",
        api_key=os.environ["PHASEO_API_KEY"],
        base_url="https://api.phaseo.app/v1",
    ),
)
```

## Node.js

```bash theme={null}
pnpm add @livekit/agents @livekit/agents-plugin-openai
```

```ts theme={null}
import { voice } from "@livekit/agents";
import * as openai from "@livekit/agents-plugin-openai";

const session = new voice.AgentSession({
	llm: new openai.LLM({
		model: "openai/gpt-5-nano",
		apiKey: process.env.PHASEO_API_KEY,
		baseURL: "https://api.phaseo.app/v1",
	}),
});
```

## Notes

* Use the OpenAI-compatible Chat Completions mode for custom endpoints.
* LiveKit's Responses-specific OpenAI tools do not automatically become Phaseo server tools.
* Configure speech-to-text and text-to-speech separately unless the selected LiveKit interface supports an OpenAI-compatible custom URL for that modality.
* Test interruption handling, tool calls, and streaming latency before production voice traffic.

## Related

* [OpenAI SDK](./openai-sdk.mdx)
* [Streaming](../streaming.mdx)
* [Tool Calling](../tool-calling.mdx)