FrontierThe story, in brief
Stability AI launches Stable Audio 3.0 with up to six-minute tracks and open weights
Open weights. Six-minute tracks. Stability AI just reset the audio model benchmark.
The KeyNews take
Why it matters
Stability AI's Stable Audio 3.0 represents a significant capability jump in generative audio—longer outputs and open-weight releases signal a strategic shift toward commoditizing audio synthesis, directly competing with closed models while enabling downstream builders.
The key facts
4 to knowStable Audio 3.0 generates music tracks up to 6 minutes long
Three models ship with open weights
Training data entirely licensed
Released May 20, 2026
Go to the source
The Decoderthe-decoder.com
Publisher excerpt: Stability AI has unveiled Stable Audio 3.0, a new generation of audio models - three of which ship with open weights. The models generate music tracks up to six minutes long and were trained entirely on licensed data, according to the company.