Stability AI Stable Audio 3 API

model·hosted audio generation api·active

Asynchronous hosted API for Stable Audio 3 text-to-audio, audio-to-audio and inpainting.

Verified facts

Official site	https://platform.stability.ai/docs/api-reference ↗
Geography	unknown
model parameter	stable-audio-3
inputs	text; optional audio
outputs	MP3; WAV; up to 380 seconds at 44.1 kHz stereo
endpoint	POST /v2beta/audio/stable-audio/text-to-audio
operator	Stability AI
functions	text-to-audio; audio-to-audio; audio inpaint; result polling
evaluation	unknown
api version	v2beta
restrictions	26 credits per successful generation at snapshot
access method	metered API
documentation	https://platform.stability.ai/docs/api-reference
authentication	Bearer API key
models exposed	stable-audio-3
safety provenance	Content moderation may reject requests
material us relevance	Commercial developer API offered in the US.
license and commercial terms	API terms and credit pricing apply.
documented training data and licensing claims	Stability AI states licensed AudioSparx and Freesound data were used.