# pipecat-ai/pipecat

Open Source framework for voice agents, multimodal apps, and realtime AI. Maintained by Daily and the community.

Repository: https://github.com/pipecat-ai/pipecat
Canonical: https://ross.abutalabs.com/products/pipecat
Homepage: https://pipecat.ai
Language: Python
License: BSD-2-Clause
License Family: permissive
Topics: ai, real-time, voice, voice-assistant, chatbot-framework, chatbots
Last push: 2026-08-26T23:32:34+00:00

## Health v2 (maintenance only)
Score: 89/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 70
- inputs: {"age_days": 980, "days_push": 7, "days_rel": 7, "gap_med": 8, "n_releases_24m": 78}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 14769, forks 2556 (observed 2026-08-28T04:11:07.644924+00:00)

## What it is
Pipecat is an open-source Python framework (BSD-2) for building real-time voice and multimodal conversational AI agents. It orchestrates speech recognition, LLMs, text-to-speech, and transports like WebRTC, WebSocket, SIP, and PSTN into composable, ultra-low-latency pipelines with support for single agents and distributed multi-agent systems.

## Use cases
- build a voice assistant that talks to users in a browser
- create a real-time AI phone agent over SIP or PSTN
- build a multimodal agent that sees and hears via video and audio
- orchestrate multiple AI agents that hand off or fan out in parallel
- build a customer support or intake voice bot
- create an AI companion or meeting assistant
- stream speech-to-text, LLM, and text-to-speech with low latency
- swap between 150+ AI providers for STT, LLM, and TTS in one pipeline

## When to choose
- you need real-time, low-latency voice or video AI agents in Python
- you want pluggable pipelines that work with many AI providers
- you need interruption handling, turn detection, and natural conversation flow
- you want client SDKs for web, mobile, and native platforms
- you plan to scale from a single bot to distributed multi-agent systems

## When to avoid
- you only need offline batch speech transcription or TTS without real-time interaction
- you need a non-Python language for the server-side agent logic
- you want a fully managed turnkey voice bot with no code
- your use case is simple text-only chat without audio or video

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, speech-recognition, tts, llm-inference, websocket, chatbot, rag, streaming, middleware, cli
- domain: artificial-intelligence, chatbots, speech-processing, large-language-models, web-development, developer-tools
- platform: python, cross-platform, self-hosted, cloud
- tags: voice-agents, realtime-ai, multimodal, webrtc, sip, voice-assistant, pipeline-orchestration, multi-agent, stt, tts, conversational-ai, ai-agents, real-time, web-server, docker

## Member repositories
- pipecat-ai/pipecat (main) score 89

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:07.644924+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 2026-08-29T17:07:12.009841+00:00, confidence not recorded.
  - readme: https://github.com/pipecat-ai/pipecat (fetched 2026-08-28T04:11:07.644924+00:00, sha 3c71f3bae018)
  - homepage: https://pipecat.ai (fetched 2026-08-29T08:05:19.757403+00:00, sha 69e50d87e99b)
  - site_page: https://docs.pipecat.ai (fetched 2026-08-29T08:05:19.766486+00:00, sha 771c85918628)
  - site_page: https://docs.pipecat.ai/pipecat/get-started/quickstart (fetched 2026-08-29T08:05:19.768298+00:00, sha b6a80e0be3f7)
- Data as of 2026-08-30T08:39:29.467469+00:00.