Pipecat
Open-source Python framework for building real-time voice and multimodal conversational AI agents.
Liked as an open-source framework for building voice agents123, where you can plug in almost any STT, LLM or TTS provider, including local models via Ollama or Nvidia models on a single GPU.456 The most common caveat is that it takes tinkering to set up7, and production plumbing like tracing and AI-to-AI test loops is still left to you.8
Why people recommend it
Pluggable providers · 3
Commenters say you can swap in nearly any STT, LLM or TTS service with little code.
“I think pipecat is the way to go. You can plug in pretty much any STT/LLM/TTS you want and go.”
“Pipecat already supports that natively, so this can be done easily with ollama.”
Show 1 more
“pipecat to glue everything together”
Local models · 3
It supports local models, including Ollama and Nvidia models running on a single consumer GPU.
“Pipecat already supports that natively, so this can be done easily with ollama.”
“There was a great post the other day showing low latency end to end using Nvidia models on a single GPU with pipecat”
Show 1 more
“pipecat to glue everything together”
Real voice agents · 3
Commenters use it for working voice agents, from daily coding assistants to a banking voice app.
“I literally have voice agents via claude agent sdk + Pipecat”
“since publishing we have swapped to Pipecat”
Show 1 more
“Pipecat voice pipeline”
Low latency · 1
One commenter points to a demo of low-latency end-to-end voice using Nvidia models on one GPU.
“There was a great post the other day showing low latency end to end using Nvidia models on a single GPU with pipecat”
Active demos · 1
The team is noted for frequently putting out interesting demos.
“Are you familiar with Pipecat https://github.com/pipecat-ai/pipecat - they put out interesting demos frequently”
Caveats and dissent
Needs tinkering · 1
Setup takes some tinkering, and no turnkey preconfigured local install is known to the commenter.
“I think pipecat is the way to go. You can plug in pretty much any STT/LLM/TTS you want and go.”
Missing plumbing · 1
One team found variable extraction, LLM call tracing and AI-to-AI testing still painful to build.
“when we switched to OSS stacks (Pipecat, LiveKit), we realise that even with great OSS, the plumbing was still painful”
Speech-to-speech · 1
One commenter has not yet gotten local speech-to-speech setups working with it.
“I am currently playing with pipecat”
Often compared with
Who recommends it in their own words
Threads where it came up
- Ask HN: What's the current best local/open speech-to-speech setup? 4 mentions
- Ask HN: Anyone using dictation with coding agents? 1 mention
- Ask HN: Who is using MCP in production? 1 mention
- Claude Is Down 1 mention
- Statecharts: hierarchical state machines 1 mention
- Show HN: PearlOS –An open source OS companion that learns and evolves around you 1 mention
Data: pipecat.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 11
“Are you familiar with Pipecat https://github.com/pipecat-ai/pipecat - they put out interesting demos frequently”
“point any voice agent platform we choose -- eleven labs, vapi, pipecat, whatever”
“I literally have voice agents via claude agent sdk + Pipecat”
“since publishing we have swapped to Pipecat”
“Pipecat voice pipeline”
“There was a great post the other day showing low latency end to end using Nvidia models on a single GPU with pipecat”
“pipecat to glue everything together”
“I am currently playing with pipecat”
“I think pipecat is the way to go. You can plug in pretty much any STT/LLM/TTS you want and go.”
“Pipecat already supports that natively, so this can be done easily with ollama.”
“when we switched to OSS stacks (Pipecat, LiveKit), we realise that even with great OSS, the plumbing was still painful”