Stability AI: New AI Models Generate 6-Min Music Tracks

Stability AI has launched new audio models that can generate music tracks over six minutes long. This new family of four AI music models, called Stable Audio 3.0, is trained entirely on licensed data. Three of these four models are open-weight, meaning they are free to download and build upon. This launch represents a significant upgrade from the previous version, Stable Audio 2.0, which had a maximum generation length of three minutes. Stability AI states that all Stable Audio 3.0 models are trained on fully licensed data. Under their community license, users own their outputs and can freely distribute and commercialize them. This development could change how music is created and shared.