Pipecat models
Which Pipecat models can you run on Modal?
Pipecat
PhoneLLM Max Alpha 2
Pipecat’s open-weight model for voice agents, with brief “microthinking” reasoning for accurate tool calling and multi-turn conversations.
TextPipecat
PhoneLLM Mini Alpha 2
Pipecat’s compact open-weight model for voice agents, optimized for fast responses and multi-turn tool calling.
TextPipecat
PhoneLLM Alpha 1
Pipecat's open-weights LLM for voice agents: a 30B Mamba-Transformer MoE (3.5B active) fine-tuned from Nemotron 3 Nano for low-latency tool calling, 256K context, BSD 2-Clause.
Text
| Model | Architecture | Context | Modalities | Input / Output |
|---|---|---|---|---|
| PhoneLLM Max Alpha 2 | 120B MoE · 12B active | 1M | Text | Dedicated only |
| PhoneLLM Mini Alpha 2 | 30B MoE · 3.5B active | 256K | Text | Dedicated only |
| PhoneLLM Alpha 1 | 30B MoE · 3.5B active | 256K | Text | Dedicated only |
Dedicated Endpoint compute is billed by the second. See compute pricing.
Latest updates

Scaling Intelligence: A landmark commission and exhibition

Sandbox resourcing: burst, don't reserve
Why you're paying for capacity you aren't using, and how to pay only for what you use.

Modal becomes a gVisor maintainer
Modal runs upwards of 40 million gVisor containers every day. Here's why we bet on it, what we've built into it, and where we want to take it next.
