# livekit/agents

A framework for building realtime voice AI agents 🤖🎙️📹

Repository: https://github.com/livekit/agents
Canonical: https://ross.abutalabs.com/products/livekit-agents
Homepage: https://docs.livekit.io/agents
Language: Python
License: Apache-2.0
License Family: permissive
Topics: ai, real-time, voice, video, agents, openai
Last push: 2026-08-26T23:31:13+00:00

## Health v2 (maintenance only)
Score: 90/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 74
- inputs: {"age_days": 1049, "days_push": 7, "days_rel": 6, "gap_med": 0, "n_releases_24m": 318}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 13176, forks 3607 (observed 2026-08-28T04:11:02.419821+00:00)

## What it is
LiveKit Agents is an open-source framework for building realtime, multimodal voice AI agents that run as programmable participants in LiveKit WebRTC rooms. It provides pluggable STT/LLM/TTS pipelines, turn detection, job scheduling, telephony integration, MCP tool support, and a built-in test framework.

## Use cases
- build a realtime voice assistant
- create an AI phone call agent
- build a customer service voice bot
- stream speech-to-text and text-to-speech in an agent pipeline
- add a voice interface to an LLM chatbot
- build a multimodal agent that can see and hear
- run voice agents on my own servers

## When to choose
- you need low-latency, realtime voice or video AI agents over WebRTC
- you want to mix and match STT, LLM, and TTS providers in one pipeline
- you need telephony (SIP) integration for inbound/outbound calls
- you want an open-source, self-hostable voice agent stack with job scheduling and testing built in

## When to avoid
- you only need text-based chatbots without realtime audio or video
- you want a fully managed, no-code solution rather than writing Python or Node.js code
- your project doesn't use or can't deploy LiveKit's WebRTC infrastructure

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, speech-recognition, tts, llm-inference, websocket, rpc, testing, chatbot
- domain: large-language-models, speech-processing, developer-tools
- platform: python, cross-platform, self-hosted, cloud
- tags: voice-ai, webrtc, realtime-agents, telephony, stt, turn-detection, mcp, multimodal, ai-agents, real-time, voice, nodejs, docker

## Member repositories
- livekit/agents (main) score 90

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:02.419821+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 2026-08-29T17:13:28.063127+00:00, confidence not recorded.
  - readme: https://github.com/livekit/agents (fetched 2026-08-28T04:11:02.419821+00:00, sha d7854f2a8d43)
  - homepage: https://docs.livekit.io/agents (fetched 2026-08-29T08:08:50.267115+00:00, sha 7ae40a5266e1)
  - site_page: https://docs.livekit.io/ (fetched 2026-08-29T08:08:50.269959+00:00, sha 89d2d6238042)
  - site_page: https://docs.livekit.io/agents/start/voice-ai (fetched 2026-08-29T08:08:50.271868+00:00, sha f1f9f41ab478)
  - site_page: https://docs.livekit.io/agents/integrations/cerebras (fetched 2026-08-29T08:08:50.273956+00:00, sha 24d61b9479f1)
  - site_page: https://docs.livekit.io/reference/python/livekit/agents/index.html (fetched 2026-08-29T08:08:50.275949+00:00, sha 5c0fbb0e93aa)
  - site_page: https://docs.livekit.io/agents/build (fetched 2026-08-29T08:08:50.283328+00:00, sha 7ae40a5266e1)
- Data as of 2026-08-30T08:39:29.467469+00:00.