Stable Audio 3
last releaseMay 2026
goblin vibe check:
stability's audio line is worth tracking separately from its image-model drama
stability ai's latest audio generation model for music, sound design, and production-adjacent audio creation. it deserves a separate catalog row because audio generation is a different creator workflow from stable diffusion image work, especially for trailers, prototypes, sfx sketches, and music beds.
generates music and audio from text promptstargets sound design, music beds, and creator production workflowsseparate audio model family from Stable Diffusion image modelsavailable through Stability AI's audio product and platform surfaces
key features
generates music and audio from text promptstargets sound design, music beds, and creator production workflowsseparate audio model family from Stable Diffusion image modelsavailable through Stability AI's audio product and platform surfaces
spec & usage
official Stability AI product page is the canonical product surface
hosted cloud access is the practical path for most users
limitations
public technical details for the newest generation are thinner than the image-model family
commercial rights and plan limits should be checked at generation time
scope:
audiotoolmusicsoundapicloudpaidmodel
launchMay 2026
last releaseMay 2026