Brands and agencies wanting licensed AI music and sound for ads; the catch is that prices could not be read anywhere.

Kenning's take

What it does

Stable Audio 2.5 generates music and sound from text or from existing audio, with tracks up to three minutes made in under two seconds on a GPU. It added audio inpainting, where a user uploads a clip and the model continues it. It is available on StableAudio.com, through the Stability AI API, on partner platforms such as fal, Replicate and ComfyUI, and on-premises with an enterprise licence.

  • Tracks up to three minutes in under two seconds on a GPU
  • Multi-part compositions with intro, development and outro
  • Audio inpainting from an uploaded clip
  • Text-to-audio and audio-to-audio
  • Trained on fully licensed data

Who it suits. Brands, agencies and creative teams that need custom music and sound for ads, games and products.

Key facts

Category
AI models
Made by
Stability AI
Based in
United Kingdom
Released
10 September 2025
How to get it
on StableAudio.com, through the Stability AI API, on fal, Replicate and ComfyUI, and on-premises with an enterprise licence
Works on
Web, API
In the UK
Yes

FAQ

When was Stable Audio 2.5 released?

Stability AI announced it on 10 September 2025.

Where can I use Stable Audio 2.5?

On StableAudio.com, through the Stability AI API, on fal, Replicate and ComfyUI, or on-premises with an enterprise licence.

How long can Stable Audio 2.5 tracks be?

Up to three minutes.

Facts from the maker's own pages and the sources listed. Checked 2026-10-11.