Google released the music model Lyria 3.5 in the Gemini app and via API. According to The Decoder, the model is also available in Flow Music, AI Studio, and Google Vids.
Google states that Lyria 3.5 was trained only on licensed content. The company also links the model to more expressive vocals and richer arrangements.
The practical impact depends on how noticeable these claimed improvements are to users and how API access is arranged. The material presented does not include details on pricing, quotas, regional availability, or terms of use for created music.
editorial commentary
Why it matters
Likely consequence is broader access to music generation within Google's ecosystem. The next observable signals will be API availability and independent assessments of vocals and arrangements. Substantial uncertainty remains about regions, pricing, limits, and rights to the results.