As of: 2026-11-13 13:49 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/m/stepaudio-3-realtime/

# StepAudio 3 Realtime

StepFun, StepAudio 3 family. Status: preview. Type: speech and audio. Released: 2026-09-15.

## How to call it

| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| StepFun API (Realtime WebSocket) | `stepaudio-3-realtime-preview` |  | https://platform.stepfun.ai/docs/en/guides/models/stepaudio-3-realtime |
| StepFun API (Chat Completions) | `stepaudio-3-chat-preview` |  | https://platform.stepfun.ai/docs/en/guides/models/stepaudio-3-realtime |

## Pricing

- Input: $0 per 1M tokens
- Output: $0 per 1M tokens
- Unit 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: https://platform.stepfun.ai/docs/en/pricing/details
- Last checked: 2026-09-29

## Specs

- Input: audio, text
- Output: audio, text
- Context window: not published
- Max output: not published
- Open weights: no
- Licence: proprietary

## Where it stands

- Newer speech and audio models in the registry since its release: 25. The 5 newest:
  - NVIDIA Nemotron 3.5 Transcribe (30B-A4B) (2026-10-28) https://postcutoff.com/m/nemotron-3-5-transcribe/
  - Hy-SimulMT-30B-A3B (2026-10-22) https://postcutoff.com/m/hy-simulmt-30b-a3b/
  - Eleven Realtime v1 Max / Mini (2026-10-20) https://postcutoff.com/m/eleven-realtime-v1-max/
  - Voxtral Mini Transcribe 3 (+ Mini Transcribe Realtime 3) (2026-10-20) https://postcutoff.com/m/voxtral-mini-transcribe-3/
  - Voxtral Small Transcribe 3 (2026-10-20) https://postcutoff.com/m/voxtral-small-transcribe-3/
- Before it from StepFun: StepFun Step-Audio-EditX (2025-11-06) https://postcutoff.com/m/step-audio-editx/
- Closed weights; access is hosted only.

## What stands out

- 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: https://arxiv.org/abs/2609.14005
- #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: https://x.com/StepFun_ai/status/2099916376274313630

## 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).

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

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

Model page: https://postcutoff.com/m/stepaudio-3-realtime/