From Walkie-Talkie to Telephone: full-duplex voice AI in H1 2026

Speech-to-speech APIs


Grok Voice 3.0 xAI	66.2%	28	2 of 5	–	$0.08/min	16
gpt-realtime-2.1 OpenAI	49.4%	10	3 of 5	1.41 s	$64/1M out	–
Gemini 3.8 Live Google	42.1% Extended Thinking, high	30	2 of 5	–	$12/1M out	99
GPT-Live 1 OpenAI	–	31	3 of 5	0.80 s	$0.05/min	–
Eleven Realtime v1 ElevenLabs, cascaded	–	17,000+ library	1 of 5	–	$0.08/min	90+
Nova 2.5 Sonic Amazon	–	16	1 of 5	–	$12/1M out	7
Qwen-Audio 3.1 Realtime Plus Alibaba	–	13	2 of 5	1.92 s	$24/1M out	11
StepAudio 3 Realtime StepFun, preview	–	7	3 of 5	–	free preview	2

Instruction following

How faithfully the voice agent follows its instructions and policy across 213 customer-service scenarios (clean audio).

τ-Voice v2, task success

independent: EVA-Bench 2.0, τ-Voice v2

Response time

Full-Duplex-Bench v1 turn-taking latency, as reported by each vendor. Lower is better.

Not published: Gemini 3.8 Live, Grok Voice 3.0, Eleven Realtime v1, Nova 2.5 Sonic, StepAudio 3 Realtime

vendor runs: OpenAI, Alibaba paper

Cost

List price per minute of session where the vendor charges by time. Extras apply; see notes.

Priced per 1M audio tokens, input and output

Gemini 3.8 Live	$3 / $12
Nova 2.5 Sonic	$3 / $12
Qwen-Audio 3.1 Realtime Plus	$6.4 / $24
gpt-realtime-2.1	$32 / $64

StepAudio 3 Realtime: Free during the preview; paid pricing not published.

vendor pricing: GPT-Live 1, gpt-realtime-2.1, Google, xAI, ElevenLabs, Amazon, Alibaba, StepFun

Voices

Built-in voices selectable through the API, and whether you can add your own.

Eleven Realtime v1: 17,000+ library

vendor docs: GPT-Live 1, gpt-realtime-2.1, Google, xAI, ElevenLabs, Amazon, Alibaba, StepFun

Expressiveness

Documented controls over how the voice speaks. No public benchmark compares style following across these APIs yet.

	Style	Emotion	Speed	Accent	Nonverbal
GPT-Live 1		–			–
gpt-realtime-2.1		–			–
StepAudio 3 Realtime				–	–
Gemini 3.8 Live			–		–
Grok Voice 3.0					–
Qwen-Audio 3.1 Realtime Plus			–	–	–
Eleven Realtime v1	–			–	–
Nova 2.5 Sonic		–	–		–

documented not available– not documented

vendor docs: GPT-Live 1, gpt-realtime-2.1, Google, xAI, ElevenLabs, Amazon, Alibaba, StepFun

Languages

Languages the model can hold a conversation in, as the vendor states.

Not published: GPT-Live 1, gpt-realtime-2.1

vendor docs: GPT-Live 1, gpt-realtime-2.1, Google, xAI, ElevenLabs, Amazon, Alibaba, StepFun

First response, in milliseconds

Time to first audible reply. Lower is better.

Cartesia Sonic-3.6Cartesia

MoshiKyutai

160

PersonaPlexNVIDIA (ADLR)

170

Sesame CSMSesame AI

200

PolyAI Dialog-RSN-1PolyAI

300

OpenAI Realtime (gpt-realtime-2.1)OpenAI

310

TML-Interaction-SmallThinking Machines Lab

400

Amazon Nova 2 SonicAmazon / AWS

400

BlandBland AI

400

NVIDIA NemotronLabs VoiceChat 11BNVIDIA

448

ElevenLabs Agents (ElevenAgents)ElevenLabs

450

Deepgram Voice AgentDeepgram

450

native full-duplexpipeline stackother

Model releases per half-year

23 H2

24 H1

24 H2

25 H1

25 H2

26 H1

26 H2

native full-duplexpipeline stackturn-basedother

benchmark	tier	setting	updated
TurnBench	native	lab	2026-10
VideoFDB	native	live	2026-10
DuplexAct	native	lab	2026-10
MultiPartyTurnBench	native	lab	2026-10
Artificial Analysis Speech Arena	adjacent	arena	2026-10
Artificial Analysis Speech Agent Arena	native	arena	2026-10