phonellm-alpha-1

About

What phonellm-alpha-1 is

Pipecat PhoneLLM Alpha 1 Official quantization The official Pipecat NVFP4 checkpoint for NVIDIA Blackwell GPUs is pipecat-ai/phonellm-alpha-1-nvfp4. It uses the same PhoneLLM Alpha 1 weights and is released with a tested native vLLM recipe for one NVIDIA B200. All figures below are means over ten complete PhoneBench runs at temperature zero with thinking disabled. For maximum benchmark quality, use NVFP4 weights with a BF16 KV cache. See the NVFP4 model card for the complete vLLM command, hardware scope, calibration details, licenses, and immutable file manifest. The Pipecat team is pleased to announce the release of PhoneLLM Alpha 1, an open-weights model for voice agent use cases. This release is the result of our ongoing work training small, open-weights LLMs for low-latency and multi-turn agentic workloads. When paired with transcription and text-to-speech models through a framework like Pipecat, PhoneLLM can handle incoming calls for financial services, healthcare, retail, and hospitality customer service, and perform common outbound calling agent tasks. PhoneLLM runs at a fraction of the cost and latency of larger, general-purpose models, while delivering comparable performance for specific use cases. For example, PhoneLLM performs on par with GPT 5.6 Terra, but 94% cheaper and with 1,300ms faster P95 time-to-first-token. PhoneLLM is an open model, so you can run it on your own infrastructure. The model is released under the BSD license, with no commercial restrictions. We are also announcing PhoneBench v1, a benchmark which evaluates LLMs based on their suitability for phone agent use cases. In addition to accuracy and speaking style, we measure model latency and estimate per-minute runtime cost. The unique combination of benchmark accuracy, low latency, and low cost makes PhoneLLM one of the most compelling options for building a voice agent. Model specs PhoneLLM Alpha 1 is a full-parameter fine-tune of NVIDIA’s Nemotron 3 Nano 30B-A3B model, trained using the NVIDIA NeMo framework. Like Nemotron Nano, PhoneLLM is a mixture-of-experts (MoE) model, with 3.5B active parameters, allowing for high-speed inference at low cost. In our PhoneBench benchmark, Pho

Summarised from the published model card. Read the full card on the HuggingFace links below.

Specifications

What it is

How it works

How language models work

Variants

Sizes & precisions

Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys phonellm-alpha-1 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (phonellm-alpha-1 below is illustrative; you get the exact model name on deployment.)

$ curl -sS https://api.axforge.ai/v1/chat/completions \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"phonellm-alpha-1","messages":[{"role":"user","content":"Hello"}]}'

Details

Languages, data & research

Licence

Open weights

Sources

Weights & code

Explore

More language models