Stable Audio 2.5

model·generative audio model·superseded

Enterprise-oriented Stable Audio release supporting three-minute generation and inpainting.

Verified facts

Official site	https://stability.ai/stable-audio ↗
Geography	unknown
api	Stability AI API
inputs	text; audio
license	not publicly disclosed
outputs	music; sound
weights	not publicly disclosed
modalities	audio; music; sound; text
dependencies	unknown
model family	Stable Audio
access method	StableAudio.com; Stability AI API; partner platforms; enterprise on-premises
exact version	2.5
repository links	unknown
evaluation results	unknown
evidence separation	developer claims: Developer reports less-than-two-second GPU inference for tracks up to three minutes. · measured results: · independent evaluations:
developer maintainer	Stability AI
material us relevance	Publicly accessible in or materially offered to the United States market as of snapshot.
documented capabilities	text-to-audio; audio-to-audio; audio inpainting; tracks up to three minutes
products exposing model	StableAudio.com; Stability AI API; fal; Replicate; ComfyUI
commercial use conditions	Commercial service and Enterprise license terms apply.
known geographic or policy restrictions	unknown
safety watermarking disclosure provenance	Content recognition is used on uploads according to the company