Model · Music and sound

Stable Audio Open 1.0

A downloadable model that turns a short English description into up to 47 seconds of stereo audio. Best at sound effects and field recordings; weaker at music and cannot sing. Needs technical setup to run.

Go to Stable Audio Open 1.0’s website

At a glance

Open or closed
Downloadable, with conditions
You can download and run it, but the licence limits use
Weights downloadable after accepting the licence on Hugging Face; licence has a revenue cap.
source · checked 12 Oct 2026 · checked on the maker's own page
Licence
Stability AI Community License (Community licence with conditions (e.g. revenue cap))
source · checked 12 Oct 2026 · checked on the maker's own page
Can I sell what I make?
Yes, with conditions
  • Only below a revenue limit
Community License: free commercial use if you and affiliates earn under US$1M a year and register with Stability AI; above that you need a separate licence. Acceptable Use Policy applies.
source · checked 12 Oct 2026 · checked on the maker's own page
Cost
Free option: Free to download and run yourself under the Community License
Plans below.
How you use it
On your own computer, Through another service
Gated: log in and accept the licence on Hugging Face. Run with stable-audio-tools or diffusers (StableAudioPipeline). No GPU memory requirement stated. Platforms inferred from Python tooling.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial
Skill needed
Technical
source · checked 12 Oct 2026 · checked, but the page was unclear or partial
Made by
Stability AI
No company location stated on the pages read.
source · checked 12 Oct 2026 · checked on the maker's own page
Status
Available
Paper published July 2024; no deprecation notice on the model card.
source · checked 12 Oct 2026 · checked on the maker's own page

What you can make with it

text to audio

Model card: better at sound effects and field recordings than music; no realistic vocals; English prompts. 'background' covers short instrument riffs and loops. Trained on CC-licensed Freesound and Free Music Archive audio.
source · checked 12 Oct 2026 · checked, but the page was unclear or partial

Plans and prices

No paid plans listed.

Enterprise licence (over US$1M revenue) is custom-priced; not listed.
source · checked 12 Oct 2026 · checked on the maker's own page

Rights and safety

Limits
max audio seconds: 47
44.1 kHz stereo output.
source · checked 12 Oct 2026 · checked on the maker's own page

Related

Where you can run it

Hugging Face

Other options for “Music and sound, sound effects”

Adobe Firefly, AudioGen (medium), Bark, Beatoven.ai, Captions, ElevenLabs, Filmora, FlexClip, getimg.ai, ImagineArt, Magic Hour, Magnific, MMAudio, MyEdit, Scenario, Stable Audio, Stable Audio 2.5

About this entry

Facts last checked 12 Oct 2026 against the maker’s own pages (each fact links to its source). Entry status: draft. Plain-English summary written by Plainly.