LiveKit Agents

What it does

LiveKit Agents is an open-source framework for building realtime voice and video AI agents in code, with Python and TypeScript libraries. Developers define agent logic, tasks, workflows, and tool use in code, then connect a speech-to-text, language model, and text-to-speech pipeline, or a realtime speech-to-speech model. The output is a server-side agent that joins LiveKit rooms and talks with end users over audio and video.

In practice, developers write agent code, test it locally in console mode, then run it in dev mode against a LiveKit room and deploy it to LiveKit Cloud or a self-hosted LiveKit server. Distinctive features include multi-agent handoffs, function tools that call external APIs, MCP server support, semantic turn detection, and a built-in test framework that uses judges to check agent behavior.

This suits developers building voice or video agents who want code-level control and are comfortable managing Python or TypeScript projects. A LiveKit Cloud account starts free with a monthly allowance of agent session minutes; higher usage, concurrency, and inference are billed. The framework is Apache-2.0 licensed, but running models and compute incurs separate costs.

What it can do

• Build voice and video agents in Python or TypeScript with the open-source Agents SDK
• Mix and match STT, LLM, and TTS pipeline components, or use a realtime speech-to-speech model
• Define custom function tools and use MCP server tools
• Handle end-of-turn detection, noise cancellation, and interruptions with built-in models
• Run multi-agent handoffs and structured tasks in code
• Test agents with an automated test framework that includes LLM judges

Works with

• Model providers: OpenAI, Deepgram, Cartesia, and other STT/LLM/TTS providers via plugins
• MCP servers
• Telephony via LiveKit SIP and Twilio connector
• WebRTC clients via LiveKit client SDKs

Example workflows

Build a voice assistant

Input: a Python or TypeScript project with LiveKit Agents installed and LiveKit credentials. Steps: define an Agent with instructions and function tools, configure an AgentSession with STT, LLM, and TTS, run locally in console mode, then start in dev mode. Output: a runnable voice agent that answers users joining a LiveKit room.

Hand off between specialist agents

Input: a conversation flow with distinct stages such as intake and storytelling. Steps: create an intro Agent that gathers information, use a function tool to record it and return the next Agent, and generate replies in each agent's on_enter. Output: an agent session that transfers control between agents while keeping shared userdata.

Before you choose

• Agent deployments on LiveKit Cloud have plan limits on agent deployments, concurrent sessions, and included session minutes; check the pricing page before scaling.
• Model usage beyond included LiveKit Inference credits is billed per model rate.
• The framework is open source under Apache-2.0, but self-hosting requires your own compute, and model providers may charge separately.

Listed under

• Coding
• Developers
• Developer tools
• Open source
• Python
• Realtime
• Realtime agents

Get started

Start with a free account, install the Python or TypeScript Agents SDK, and follow the voice AI quickstart.

1. Install the Agents library with pip (for example livekit-agents with selected plugins) or the TypeScript SDK.
2. Set LIVEKIT_URL, LIVEKIT_API_KEY, and LIVEKIT_API_SECRET for the LiveKit server or Cloud project.
3. Run the agent locally in console mode to test, then in dev mode to connect to a LiveKit room.

The software can be installed under its repository license. Infrastructure and model-provider charges remain separate. Optional commercial hosting and support have their own terms.

LiveKit Agents is free to use. See their pricing page ↗

Sources for this overview

Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Agents SDKs | Build realtime AI agents in code | LiveKit ↗livekit.com
• GitHub - livekit/agents: A framework for building realtime voice AI agents 🤖🎙️📹 · GitHub ↗github.com
• Pricing | LiveKit ↗livekit.com

Company website

livekit.com: product, pricing and policy pages

YC directory mirror

Batch and company name

Web search

Confirms the official site

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.