Overview
Build AI agents that join LiveKit rooms as participants, process realtime media and data streams, and interact with users through voice, text, and vision. The LiveKit Agents framework provides everything you need to build production-ready voice AI agents and programmatic participants.
When you build agents with the Agents framework, they join rooms as participants just like users from frontend apps. Agents can process audio, video, and data streams in realtime, making them ideal for voice assistants, multimodal AI applications, and custom programmatic participants.
The framework allows you to add Python or Node.js programs to any LiveKit room as full realtime participants. It includes tools and abstractions that make it easy to feed realtime media and data through an AI pipeline that works with any provider, and to publish realtime results back to the room.
Getting started
Build your first agent with these resources:
Voice AI quickstart
Build and deploy a simple voice assistant with Python or Node.js in less than 10 minutes.
LiveKit Agent Builder
Prototype and deploy voice agents directly in your browser, without writing any code.
Learn more
For complete documentation on building agents:
Agents framework
Learn how to build AI agents and programmatic participants with the LiveKit Agents framework.
Multimodality
Learn how to configure agents to process speech, text, and vision inputs.
Logic & structure
Learn how to structure your agent's logic and behavior with sessions, tasks, and workflows.
Agent server
Learn how agent servers manage your agents' lifecycle and deployment.
Models
Explore the full list of AI models and providers available for your agents.