# livekit/livekit

End-to-end realtime stack for connecting humans and AI

Repository: https://github.com/livekit/livekit
Canonical: https://ross.abutalabs.com/products/livekit
Homepage: https://docs.livekit.io
Language: Go
License: Apache-2.0
License Family: permissive
Topics: golang, webrtc, sfu, media-server, video, voice, voice-ai
Last push: 2026-08-26T17:32:15+00:00

## Health v2 (maintenance only)
Score: 99/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 99, longevity 100
- inputs: {"age_days": 2163, "days_push": 7, "days_rel": 7, "gap_med": 16, "n_releases_24m": 26}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 20530, forks 2280 (observed 2026-08-28T04:11:30.189629+00:00)

## What it is
LiveKit is an open-source, scalable WebRTC SFU media server written in Go that provides realtime video, audio, and data transport for applications. It also serves as the foundation for the LiveKit Agents framework for building voice, video, and physical AI agents, with SDKs across many languages and platforms.

## Use cases
- build a multi-user video conferencing app
- create a voice assistant that talks to users in realtime
- stream live audio and video between browsers and servers
- build an AI agent that answers phone calls
- add realtime data channels to a multiplayer app
- self-host a scalable WebRTC media server
- connect a robot to realtime audio and video streams

## When to choose
- you need low-latency multi-party audio/video with a proven WebRTC stack
- you are building voice or multimodal AI agents that interact with users in realtime
- you want self-hosting or cloud deployment options with JWT auth, TURN, and multi-region scaling
- you need SDKs across web, mobile, Unity, and embedded platforms

## When to avoid
- you only need simple request-response HTTP APIs with no realtime media
- you need a fully managed turnkey video product with no infrastructure work and cannot run LiveKit Cloud
- your use case is one-way streaming where simpler protocols like HLS suffice

## Facets
- artifact type: service
- maturity: stable
- function: websocket, rpc, auth, agent-framework, chatbot, speech-recognition, tts, llm-inference, webhook, sdk
- domain: artificial-intelligence, web-development, media, robotics
- platform: go, self-hosted, cross-platform, cloud
- tags: webrtc, sfu, media-server, voice-ai, video-conferencing, realtime-communication, simulcast, end-to-end-encryption, real-time, ai-agents, telephony, docker, kubernetes, web-server

## Member repositories
- livekit/livekit (main) score 99

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:30.189629+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:58:46.909003+00:00, confidence not recorded.
  - readme: https://github.com/livekit/livekit (fetched 2026-08-28T04:11:30.189629+00:00, sha 729863086193)
  - homepage: https://docs.livekit.io (fetched 2026-08-29T07:57:26.697361+00:00, sha 89d2d6238042)
  - site_page: https://docs.livekit.io/ (fetched 2026-08-29T07:57:26.707150+00:00, sha 89d2d6238042)
  - site_page: https://docs.livekit.io/intro/about (fetched 2026-08-29T07:57:26.709474+00:00, sha 2bf82fdd060a)
  - site_page: https://docs.livekit.io/agents/start/voice-ai-quickstart (fetched 2026-08-29T07:57:26.711523+00:00, sha f1f9f41ab478)
  - site_page: https://docs.livekit.io/intro/coding-agents (fetched 2026-08-29T07:57:26.713598+00:00, sha e2bc73d4d72f)
  - site_page: https://docs.livekit.io/reference (fetched 2026-08-29T07:57:26.715318+00:00, sha 2b28b0ba6214)
  - site_page: https://docs.livekit.io/reference/recipes?tag=rag (fetched 2026-08-29T07:57:26.717218+00:00, sha 3c4da147a68a)
- Data as of 2026-08-30T08:39:29.467469+00:00.