Step-1o Audio

Name: Step-1o Audio
Author: StepFun

audioproprietary

* 定位：稳定型，久经考验 * 能力标签：Tool Call * 接入方式：仅支持 API 接入 * 第一代端到端语音模型。技术成熟稳定，在汽车场景下大规模部署，久经考验，支持多种预置音色风格，支持工具调用。适用于构建基础的语音交互、内容生成等常规业务场景。

Input price

$3.47/1M tokens

Output price

$8.33/1M tokens

Capabilities

Show unsupported (12)

Infrastructure

Streaming

Details

Model IDstep-1o-audio

ProviderStepFun

Creatorstepfun

Familystep-1o

Licenseproprietary

Statusactive

Typeaudio

Input modalitiestext, audio

Output modalitiestext, audio

Open weightNo

Parameters—

Architecture—

Max duration—

Sample rate—

Channels—

Languages—

Genres—

Release date—

Sourceofficial

Endpoints

RealtimeWS

Low-latency streaming via WebSockethttps://api.stepfun.com/v1/realtime

Pricing

Input

$3.47

Output

$8.33

Cache write

—

Cache read

$0.69

Batch in

—

Batch out

—

Family Comparison: step-1o

Model	Caps	Perf	Speed	Context	Max out	Pricing
Step-1o Turbo Vision	TS	—	—	32K	—	$0.35$1.11
Step 1o Vision 32k	S	—	—	32K	—	$2.08$9.72
Step-1o Audio	S	—	—	—	—	$3.47$8.33

API

GET/v1/models/stepfun/step-1o-audio