🎵 Stable Audio 3
Text-to-audio generation with
Stable Audio 3
. Pick a variant, write a prompt, hit Generate.
Model
Medium — general audio (largest)
Small Music — 0.6B, music-focused
Small SFX — 0.6B, sound effects
Prompt
Duration (s) · model max 380s
↺
1
380
Advanced settings
â–¼
🎼 Generate
Output
Examples (lazy-cached on first click)