---
title: pipecat-ai/phonellm-mini-alpha-2
url: https://recentmodel.com/models/pipecat-ai/phonellm-mini-alpha-2/
status: archived · low-signal
published: 2026-10-15T15:23:14Z
updated: 2026-10-22T23:27:46Z
source: https://huggingface.co/pipecat-ai/phonellm-mini-alpha-2
---

# pipecat-ai/phonellm-mini-alpha-2

PhoneLLM Mini Alpha 2 is a full-parameter supervised fine-tune of NVIDIA Nemotron 3 Nano 30B-A3B for phone voice agents with tool calling. Mixture-of-experts, 32B total (128 experts, 6 per token). Context: 256K tokens (config). License: bsd-2-clause.


- **Org:** pipecat-ai
- **Released:** 2026-10-15
- **Status:** archived · low-signal
- **License:** bsd-2-clause (permissive)
- **Parameters:** 31.6B
- **Architecture:** MoE · 128 experts, 6/token · NemotronHForCausalLM
- **Context:** 262,144 tokens
- **Task:** text-generation · text
- **Signal:** 7 HF likes · 896 downloads
- **Archived:** 2026-10-22 (no longer listed by the collector; stats as last seen)

## Runs on

real quant files; sizing calculation, not tested.

- Q4 ~25.9 GB: 32GB GPU (5090), Mac 36GB, CPU only, 32GB RAM (slow)
- Q8 ~35.0 GB: 48GB GPU (RTX 6000 Ada / A6000), Mac 48GB, Mac 64GB (M1 Max), CPU only, 64GB RAM (slow)
- FP16 ~64.6 GB: 96GB GPU (RTX Pro 6000 Blackwell), 1× DGX Spark (128GB), 2× DGX Spark (256GB), Strix Halo 128GB unified (~96GB usable), Mac 96GB, Mac 128GB, Mac 192GB, Mac 256GB, Mac 512GB, CPU only, 128GB RAM (slow), CPU only, 256GB RAM (slow)

## Links

- [Hugging Face](https://huggingface.co/pipecat-ai/phonellm-mini-alpha-2)