livekit/agents

What is it?

Key features

• 01Mix and match STT, LLM, TTS, and realtime API providers in one pipeline.
• 02Semantic turn detection using a transformer model to reduce interruptions.
• 03Native MCP support to load tools from MCP servers.
• 04Telephony integration for making and receiving phone calls.
• 05Data exchange with clients via RPCs and data APIs.
• 06Built-in test framework with judges for checking agent behavior.

Use cases

• Building a customer support voice agent that answers calls over the phone.
• Creating an outbound calling agent that places automated calls.
• Adding a multi-user transcriber to produce transcripts for everyone in a room.
• Developing a vision-enabled assistant that reacts to live video from users.

Quick start

1. 1

   Install the core library with model provider plugins

   pip install "livekit-agents[openai,deepgram,cartesia]"
2. 2

   Set LIVEKIT_URL, LIVEKIT_API_KEY and LIVEKIT_API_SECRET environment variables

3. 3

   Save the example agent as myagent.py and test it in terminal mode

   python myagent.py console

Things to know

• !The quick start example relies on LiveKit Inference or external model provider API keys, which are typically paid services.
• !Connecting to a live room requires a LiveKit Cloud account or a self-hosted LiveKit server.

Topics

FAQ

What is agents?

A Python framework for building realtime voice and video AI agents that run as server-side participants in LiveKit rooms. Agents is a Python framework for building realtime, programmable participants that run on servers. It lets developers create conversational, multi-modal voice agents that can hear, see, and respond within LiveKit rooms, connecting end users to agents through dispatch APIs.

Is agents free to use? What license does it have?

agents is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.

Is agents still maintained?

Actively maintained. The last commit to agents was on Oct 22, 2026.

What can I use agents for?

Building a customer support voice agent that answers calls over the phone. Creating an outbound calling agent that places automated calls. Adding a multi-user transcriber to produce transcripts for everyone in a room. Developing a vision-enabled assistant that reacts to live video from users.

How do I install agents?

Install the core library with model provider plugins: pip install "livekit-agents[openai,deepgram,cartesia]" Set LIVEKIT_URL, LIVEKIT_API_KEY and LIVEKIT_API_SECRET environment variables Save the example agent as myagent.py and test it in terminal mode: python myagent.py console

Where can I get agents?

The source code and installation instructions are on GitHub: https://github.com/livekit/agents