Call a customer-service line, and human patience lasts about 1.5 seconds. That is the hard number Daily — the company behind the open-source Pipecat voice-agent framework — anchored its PhoneLLM Alpha 1 release on: voice-to-voice latency needs to stay around 1,500ms for a phone conversation to feel natural. Their own measurement put GPT 5.6 Terra's P95 time-to-first-token in fast mode at about 1,900ms, meaning the LLM alone blows the budget before STT, TTS, and network overhead even enter the picture. That is the reason PhoneLLM exists: an open-weights model trained specifically for the phone.
A 30B MoE with 3.5B active parameters, built to take calls
The model card lays out the specifics: PhoneLLM Alpha 1 is a full-parameter fine-tune of NVIDIA's Nemotron 3 Nano 30B-A3B, trained with the NVIDIA NeMo framework. The architecture is a hybrid Mamba-Transformer mixture-of-experts — 30B total parameters, only 3.5B active — with a 262,144-token context window, BF16 safetensors, and a BSD 2-Clause license with no commercial restrictions. The recommended inference settings are blunt: temperature=0, thinking disabled, because that is exactly the regime the model was trained for.
The target scenarios are deliberately narrow: inbound customer service plus routine outbound calling for financial services, healthcare, retail, and hospitality. The problem it attacks is very specific: with thinking disabled, models large and small are unreliable at tool invocation across long, multi-turn conversations — an LLM will cheerfully say "Yes, I've booked that table for you" without ever calling the booking tool. PhoneLLM is trained to invoke the right tools at the right time, no chain-of-thought required.
Where $0.00025 per minute comes from
The model card walks through a very engineering-grade cost calculation. A single B200 hosts 44 concurrent agent processes (88 per two-GPU node). On Modal, a B200 costs $6.2496/hour at base; region pinning adds a 1.5x multiplier for $9.3744/hour; targeting 70% utilization yields an effective $13.392/hour, or $0.2232/minute. Divide by 88 concurrent agents and you get $0.00025 per agent-minute. Single-request TTFT P95 lands below 100ms on a B200, and with Modal AutoEndpoints' tuned configuration, maximum concurrency at the sub-600ms P95 time-to-first-audio-token target is roughly double what the generic vLLM cookbook configuration delivers.
Daily's official framing is bolder: performance on par with GPT 5.6 Terra on voice-agent tasks, at 94% lower cost and with a 1,300ms faster P95 time-to-first-token. Co-founder Kwindla Hultman Kramer put it on X as one-third the latency and one-eighteenth the cost — and the two framings are mathematically consistent (94% off is roughly 1/18th the price).
The self-authored benchmark problem
All of those numbers come from Daily's own PhoneBench v1: LLM judges calibrated against human labels, grading telephone speaking style, tool-call accuracy, say/do consistency, factual grounding, conversation coherence, authentication and escalation discipline, and caller outcome. The independent outlet explainx.ai poured the appropriate cold water: "Alpha 1" is a label the developers gave it themselves; the benchmark is self-authored, with no third-party public leaderboard reproduction; the numbers are configuration-sensitive (temperature=0 plus thinking off, or they likely won't replicate); and the comparison is a mismatched one — a specialized self-hosted deployment versus a general-purpose hosted API. Terra's latency and cost are what OpenAI's endpoint hands you with zero infrastructure work; PhoneLLM's numbers assume you operate the B200 yourself. After OpenAI's late-July price cut, GPT-5.6 Terra lists at $2/$12 per million tokens, and the two framings do not sit on the same ledger when it comes to converting that into per-minute cost.
So what
Even with the vendor's numbers cut in half, the release is worth attention. This is not "another small model" — it welds task-specific fine-tuning, low-activation MoE, and self-hosted concurrency engineering onto a single latency axis. Daily's read is that the industry is shifting toward small, open, purpose-built models, with weights iterated monthly using production agent traces and proprietary data. Phone customer service happens to be one of the harshest scenes for the triple constraint of latency, cost, and tool reliability — whether PhoneLLM holds up awaits third-party testing. But the route of "doing one specific job at 1/18th the cost" already has its first graded submission.
Model card and weights: https://huggingface.co/pipecat-ai/phonellm-alpha-1