Google brings Lyria 3.5 music generation to Gemini

Google just launched Lyria 3.5, calling it its best-sounding music generation model yet, and made it immediately available inside the Gemini app and Gemini API.

Expressive vocals, longer tracks, and built-in templates

The update focuses heavily on audio quality, delivering more expressive vocals and richer musical arrangements with higher fidelity output.

Inside the Gemini app on web and mobile, users can now pick or describe a genre, toggle between vocal and instrumental tracks, and choose between short clips or longer compositions. Google also rolled out pre-made templates designed for quick turnarounds, whether you need a background track for a video or a custom birthday song.

Beyond consumer Gemini access, Lyria 3.5 is also rolling out globally across Google's creative and developer stack. Artists can tap into it via Google Flow Music, video creators can use it inside Google Vids, and developers can access it through Google AI Studio and the Gemini API.

Why it matters

Dedicated AI audio platforms proved that people love generating tracks from text, but Google is playing an ecosystem game here.

For creators and small businesses, churning out bespoke video backing tracks or commercial jingles no longer requires tab-hopping to third-party tools. Dropping Lyria straight into Google Vids and Gemini makes music generation a native step in day-to-day content production.

We think this broad integration is the smartest part of the launch. Even if standalone tools push the boundary on pure musical complexity, Google embedding decent audio directly into everyday workflows puts serious pressure on single-purpose AI music startups.