Devtoolwatch BoardGet it every Monday

LiveKit Agents review

LiveKit Agents is an open-source Python framework for building realtime voice, video, and multimodal AI agents that join LiveKit rooms as participants.

Updated Nov 1, 2026. Review written Oct 22, 2026 from the project's own sources; usage data refreshed every Monday.

TL;DR

LiveKit Agents is a framework for building realtime, programmable voice and video agents that run on servers. It is for developers building conversational agents, phone agents, or multimodal assistants. It lets you mix STT, LLM, TTS, and realtime API providers, and it includes job dispatch, semantic turn detection, MCP support, telephony integration, and a built-in test framework.

LiveKit Agents at a glance

Category
Voice and realtime
Maintained by
LiveKit (organization, @livekit on GitHub)
License
Apache 2.0
Language
Python, C, Makefile
Repository
github.com/livekit/agents
Website
docs.livekit.io/agents
npm package
@livekit/agents 1.9.3
PyPI package
livekit-agents 1.8.8
Latest release
[email protected] on GitHub, Oct 15, 2026
First published
Oct 19, 2023

What LiveKit Agents does

  • LiveKit Agents lets you combine STT, LLM, TTS, and realtime API providers through plugins to fit your use case.
  • LiveKit Agents includes built-in job scheduling and dispatch APIs to connect end users to agents.
  • LiveKit Agents uses a transformer model for semantic turn detection, which helps reduce interruptions.
  • LiveKit Agents integrates with LiveKit's telephony stack so agents can make or receive phone calls.
  • LiveKit Agents supports MCP natively, so you can integrate tools from MCP servers with one line of code.
  • LiveKit Agents supports multi-agent handoffs, where one agent passes the session to another agent with different instructions or models.
  • LiveKit Agents includes a test framework that checks agent events and uses LLM judges to evaluate responses.

How it works

You write a Python program that defines an Agent, an AgentSession, and an entrypoint, which is run by an AgentServer. The agent joins LiveKit rooms as a realtime participant through LiveKit Cloud or a self-hosted LiveKit server. Users connect with LiveKit client SDKs or telephony. Models are called through LiveKit Inference or directly with provider keys.

How to install and start using LiveKit Agents

With pip (core plus common plugins)

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

Requirements

  • Python 3.10 or later and below 3.15.
  • LiveKit server credentials (LIVEKIT_URL, LIVEKIT_API_KEY, LIVEKIT_API_SECRET) for LiveKit Cloud or a self-hosted server.
  • API keys for model providers if you call them directly instead of through LiveKit Inference.

First steps

  1. Install LiveKit Agents with the plugins for your model providers using pip.
  2. Set the LIVEKIT_URL, LIVEKIT_API_KEY, and LIVEKIT_API_SECRET environment variables.
  3. Run python myagent.py console to test the agent in your terminal with local audio input and output.
  4. Run python myagent.py dev to start the agent server with hot reloading, then connect through the Agents Playground.

Who LiveKit Agents is for

  • Developers building conversational voice assistants that run on servers.
  • Teams building phone agents that place or receive calls through LiveKit telephony.
  • Developers who want to self-host the full realtime voice stack, including the LiveKit server.
  • Multimodal agents that work with video, vision, or avatars.

Limitations to know

  • LiveKit Agents connects through LiveKit rooms, so you need LiveKit Cloud or a self-hosted LiveKit server for client connections.
  • The LiveKit turn detection models use the LiveKit Model License, not Apache-2.0.
  • This repository is the Python library. JavaScript and TypeScript developers use the separate AgentsJS library.
  • The framework is under active development, and recent releases have included breaking changes.

Is LiveKit Agents growing? Usage data

In the week of Oct 25–31, 2026, LiveKit Agents ranked #122 on the Devtoolwatch Board and #7 in Voice and realtime, with a momentum score of 58 out of 100. Its main package had about 2.4M downloads that week, +14% against its own 4-week average.

New GitHub stars per week for LiveKit Agents12 weeks from Aug 9, 2026: 253, 113, 443, 468, 140, 122, 99, 114, 228, 125, 133, 105 new stars, an average of 195 a week.Aug 9week of Oct 25
New GitHub stars per week for the last 12 weeks, from GitHub's stargazer history (today's stargazers by the week they starred).
Show the weekly numbers
New GitHub stars per week for LiveKit Agents
Week ofNew stars
Aug 9, 2026253
Aug 16, 2026113
Aug 23, 2026443
Aug 30, 2026468
Sep 6, 2026140
Sep 13, 2026122
Sep 20, 202699
Sep 27, 2026114
Oct 4, 2026228
Oct 11, 2026125
Oct 18, 2026133
Oct 25, 2026105

On Hacker News

The Board ranks tools by four signals each week: new stars, new stars for the repo's size, added downloads and download growth beyond the market. How the ranking works.

LiveKit Agents alternatives

Other voice and realtime AI tools on the Board in the week of Oct 25–31, 2026, by momentum:

Voice and realtime ranked by momentum, the week of Oct 25–31, 2026
RankToolStarsDownloads/wk
#1faster-whisperfaster-whisper is a Python reimplementation of OpenAI's Whisper speech-to-text model built on the CTranslate2 inference engine for…26.4k4.4M
#2KokoroKokoro is a Python inference library for Kokoro-82M, an open-weight text-to-speech model with 82 million parameters.9.8k150.7k
#3ChatterboxChatterbox is a family of open-source text-to-speech and voice cloning models by Resemble AI, with English and multilingual variants.27.5k73.2k
#4VoiceMemInfrastructure for the next generation of voice agents, designed to provide universal memory. It is divided into a left brain and a…2.8k–
#5PipecatPipecat is an open-source Python framework for building real-time voice and multimodal conversational agents, maintained by Daily and…17k348.5k
#6WhisperXWhisperX is a Python speech recognition tool built on Whisper that adds word-level timestamps and speaker diarization.25k158.6k

All voice and realtime AI tools, ranked

Recent LiveKit Agents releases

LiveKit Agents FAQ

Is LiveKit Agents free and open source?

Yes. LiveKit Agents's source code is on GitHub (livekit/agents) under the Apache 2.0 license, so you can use it for free.

How do I install LiveKit Agents?

Run pip install "livekit-agents[openai,deepgram,cartesia]". The install section above has the other options from the project's README.

Is LiveKit Agents still maintained?

Yes. The livekit/agents repository was last updated on Oct 22, 2026, and the latest release, [email protected] on GitHub, came out on Oct 15, 2026.

How popular is LiveKit Agents?

LiveKit Agents has 15k GitHub stars as of Nov 1, 2026. It gained 105 new stars in the week of Oct 25–31, 2026, against a 12-week average of 195 a week. Its main package had about 2.4M downloads that week, +14% against its own 4-week average. That made it #7 among voice and realtime AI tools on the Devtoolwatch Board.

LiveKit Agents vs Pipecat: how do they compare?

LiveKit Agents and Pipecat are both tracked by Devtoolwatch as voice and realtime AI tools. LiveKit Agents has 15k GitHub stars and Pipecat has 17k. In the week of Oct 25–31, 2026, LiveKit Agents's main package had 2.4M downloads and Pipecat's had 348.5k. Pipecat is an open-source Python framework for building real-time voice and multimodal conversational agents, maintained by Daily and the community. Pipecat review.

What are the best alternatives to LiveKit Agents?

Other voice and realtime AI tools on the Devtoolwatch Board in the week of Oct 25–31, 2026, by momentum: faster-whisper (#1), Kokoro (#2), Chatterbox (#3), VoiceMem (#4), Pipecat (#5). The alternatives table above has their numbers.

What is LiveKit Agents?

LiveKit Agents is a framework for building realtime, programmable participants that run on servers. It adds Python or Node.js programs to LiveKit rooms as realtime participants, so you can build voice and multimodal agents that see, hear, and understand.

Can I test a LiveKit Agents agent without a server?

Yes. Running python myagent.py console starts the agent in terminal mode with local audio input and output. This mode does not require external servers or dependencies.

Do I need model provider API keys to use LiveKit Agents?

No, not always. LiveKit Inference, through LiveKit Cloud, runs models with your LiveKit credentials and no separate provider keys. You can also use provider plugins with your own keys.

Can LiveKit Agents be self-hosted?

Yes. LiveKit Agents is open source, and you can run the entire stack on your own servers, including the LiveKit server. You can also deploy agents to LiveKit Cloud or any custom environment.

Does LiveKit Agents support Node.js?

Yes. A Node.js version is available as the @livekit/agents npm package, also called AgentsJS. The main repository is the Python library.

Sources

  1. livekit/agents on GitHub
  2. LiveKit Agents README
  3. @livekit/agents on npm
  4. livekit-agents on PyPI
  5. LiveKit Agents website
  6. Release [email protected] (Oct 15, 2026)
  7. Release [email protected] (Oct 14, 2026)
  8. Release [email protected] (Oct 9, 2026)
  9. Hacker News: Show HN: Open source framework OpenAI uses for Advanced Voice
  10. Devtoolwatch Board, the week of Oct 25–31, 2026

LiveKit Agents README badge

Maintainers can show the tool's weekly Board rank in their README:

Devtoolwatch badge for LiveKit Agents

[![Devtoolwatch](https://index.devtoolwatch.com/badge/livekit-agents.svg)](https://index.devtoolwatch.com/tools/livekit-agents/)

How this page is made: Devtoolwatch writes each review from the project's README, releases, package metadata and website, with an AI model, then checks every claim against the source it cites and drops what the sources don't support. The numbers come from Devtoolwatch's daily GitHub snapshots, GitHub's stargazer history, and the npm and PyPI download APIs. Spotted an error? Email [email protected].