Agora

We recommend using Tavus’s Full Pipeline in its entirety for the lowest latency and most optimized multimodal experience. Third party integrations only provide rendering, while our Full Pipeline includes perception, turn-taking, and rendering for complete conversational intelligence.

Agora runs your conversational stack (ASR, LLM, TTS, and the RTC channel) through its Conversational AI Engine. Tavus supplies the face’s synchronized video and audio as the officially supported TavusAvatar vendor (vendor: "tavus"), joining the RTC channel alongside your agent. For Agora’s side of the setup, see Agora’s official Tavus avatar documentation.

Prerequisites

Before integrating Tavus with Agora, ensure you have the following:

Tavus API Key
Tavus face_id (or a pal_id)
- avatar_id accepts either a face_id or a pal_id — pass whichever one identifies the face/PAL you want Agora to render.
- You can use one of Tavus’s stock faces or your own custom face.
- We recommend using Phoenix-4.5 Pro faces, which are optimized for low-latency, real-time applications.
An Agora Conversational AI Agent
- An Agora project with the Conversational AI Engine enabled, and an agent already configured with your ASR, LLM, and TTS vendors.
- Follow Agora’s Conversational AI Engine quickstart if you don’t have one yet.

Integration Guide

Agora already knows Tavus’s api_base_url, so it isn’t a parameter here.

TavusAvatar parameter	Required	Description
`api_key` / `apiKey`	Yes	Your Tavus API key
`avatar_id` / `avatarId`	Yes	Your Tavus `face_id` or `pal_id`
`agora_uid` / `agoraUid`	Yes	The RTC user ID the Tavus avatar joins the channel as
`agora_token` / `agoraToken`	No	The RTC token that authorizes the avatar to join the video channel. Generate this token using your Agora project credentials

Step 1: Setup and Authentication

Generate a Tavus API key from the PAL Maker. See Authentication for how Tavus APIs use x-api-key.

Step 2: Configure Tavus Avatar

Python SDK
TypeScript
Go
REST API

from agora_agent import Agent
from agora_agent.agentkit.vendors import Tavus

# client is your configured Agora client
agent = (
    Agent(client)
    .with_stt(...)  # configure your STT vendor
    .with_llm(...)  # configure your LLM vendor
    .with_tts(...)  # configure your TTS vendor
    .with_avatar(Tavus(
        api_key="<TAVUS_API_KEY>",
        avatar_id="<TAVUS_FACE_ID_OR_PAL_ID>",
        agora_uid="<AGORA_UID>"
    ))
)

import { Agent, Tavus } from 'agora-agents';

// client is your configured Agora client
const agent = new Agent({ client })
  .withStt(/* configure your STT vendor */)
  .withLlm(/* configure your LLM vendor */)
  .withTts(/* configure your TTS vendor */)
  .withAvatar(new Tavus({
    apiKey: '<TAVUS_API_KEY>',
    avatarId: '<TAVUS_FACE_ID_OR_PAL_ID>',
    agoraUid: '<AGORA_UID>',
  }));

import (
    "github.com/AgoraIO/agora-agents-go/v2/agentkit"
    "github.com/AgoraIO/agora-agents-go/v2/agentkit/vendors"
)
// client is your configured Agora client
agent := agentkit.NewAgent(client).WithStt(/* configure your STT vendor */).
    WithLlm(/* configure your LLM vendor */).
    WithTts(/* configure your TTS vendor */).
    WithAvatar(
        vendors.NewTavus(vendors.TavusOptions{
            APIKey:   "<TAVUS_API_KEY>",
            AvatarID: "<TAVUS_FACE_ID_OR_PAL_ID>",
            AgoraUID: "<AGORA_UID>",
        }),
    )

{
  "avatar": {
    "vendor": "generic",
    "enable": true,
    "params": {
      "api_key": "<TAVUS_API_KEY>",
      "api_base_url": "https://tavusapi.com/v2/conversations/agora",
      "avatar_id": "<TAVUS_FACE_ID_OR_PAL_ID>",
      "agora_uid": "<AGORA_UID>"
    }
  }
}

Step 3: Start the agent

Start (or join) your Agora Conversational AI agent as usual. The Tavus face joins the RTC channel as agora_uid and renders synchronized video and audio driven by your agent’s TTS output.

Getting Started

Guides

Build

Deploy

Debug

Teams & SSO

Resources

Prerequisites

Integration Guide

​Prerequisites

​Integration Guide

Prerequisites

Integration Guide