Step TTS 2
by StepFun (Global) · family step · listed Mar 2026
Speech generation model for controllable voice, narration, and audio delivery
Current rates USD per 1M tokens
Input—
Output—
Cache read—
Cache write—
Against the market
On the record
- Model id
step-tts-2- Context window
- — tokens
- Max output
- —
- Input modalities
- text
- Output modalities
- audio
- Reasoning
- no
- Tool calls
- no
- Open weights
- no
- Knowledge cutoff
- —
- Released
- 2026-03-01
- Last updated
- 2026-07-02
- Provider docs
- platform.stepfun.ai/docs/en/overview/concept
Also listed at
- StepFun (China)— in · — out