Post-Cutoff
Search For AI agents

StepFunStepAudio 3 familyPreviewSpeech and audioReleased 15 Sep 2026

StepAudio 3 Realtime

StepFun’s full-duplex voice agent model; part of the five-model StepAudio 3 family (Realtime, ASR Max, TTS, Gen, Music).

Model idstepaudio-3-realtime-preview

PriceFree

Checked29 September 2026

Where it stands

25newer speech and audio models since its release

The 5 newest are listed. All speech and audio models

Before it from StepFun
StepFun Step-Audio-EditX6 Nov 2025
Weights
Closed weights; access is hosted only.

How to call it

2 ways to use StepAudio 3 Realtime, with the model id each provider expects.

Providers, model ids, endpoints and docs for StepAudio 3 Realtime
ProviderModel idEndpointDocs
StepFun API (Realtime WebSocket) stepaudio-3-realtime-preview platform.stepfun.ai
StepFun API (Chat Completions) stepaudio-3-chat-preview platform.stepfun.ai

Pricing

Prices for StepAudio 3 Realtime
Input$0 per 1M tokens
Output$0 per 1M tokens

As published: “free during limited-time preview (successor stepaudio-2.5-realtime: $1.50 in / $0.30 cached / $10.00 out per 1M tokens)”.

Source: platform.stepfun.ai, checked 29 September 2026.

Specs

Input
Audio and text
Output
Audio and text
Context window
Not published
Max output
Not published
Open weights
No
Licence
Proprietary
Released
15 September 2026
Status
Preview
Type
Speech and audio

What stands out

What its maker marketed at launch, and what people found later, each with its source.

  • Think-while-speaking full duplex

    Runs private chain-of-thought in parallel with spoken output; distinguishes real interruptions from backchannels; asynchronous tool execution (web search, knowledge retrieval).

    Source: arxiv.org

  • #1 on Artificial Analysis conversational dynamics

    98.9 on Artificial Analysis Full-Duplex Bench (Conversational Dynamics) and 99.7% Speech Reasoning at launch, ahead of Qwen Audio 3.0 Realtime Plus and GPT-Live-1 per StepFun.

    Source: x.com

Notes

Chinese and English. Preview ids will be retired for a paid GA version when the trial ends. Predecessor stepaudio-2.5-realtime (2026-05-26; persona role-play, project page https://stepaudiollm.github.io/step-audio-2.5-realtime/ with self-reported 86.36 general dialogue / 79.80 spoken QA / 82.18 paralinguistics, claimed to beat GPT-Realtime-1.5 on StepFun’s evals). Technical report arXiv 2609.14005 (56.0% task success on tau-Voice).

Sources: https://platform.stepfun.ai/docs/en/guides/models/audio · https://platform.stepfun.ai/docs/en/pricing/details · https://arxiv.org/abs/2609.14005