Google launched Lyria 3 Pro, an AI music generation model that creates tracks up to three minutes long with full song structure. The model rolls out to paid Gemini subscribers, Google Vids, and the recently acquired ProducerAI platform.

What Happened

Lyria 3 Pro is a major upgrade from the standard Lyria 3 model that Google released in February 2026. Where the original capped output at 30 seconds, Pro extends generation to three full minutes. The model understands track structure, so users can request specific intros, verses, choruses, and bridges rather than getting a single continuous loop.

Alongside the consumer launch, Google released Lyria 3 for developers via paid preview through the Gemini API and for testing in Google AI Studio. Third-party apps can now access music generation through Google's infrastructure for the first time.

Google also confirmed it acquired ProducerAI, a generative music production tool, and is integrating Lyria 3 Pro directly into the product. Google Vids, the company's AI video editing app, also gains Lyria 3 Pro for automatic soundtrack generation. The DeepMind team behind the model says the structural understanding came from training on compositions with labeled sections, letting the model learn how professional tracks are arranged.

The launch arrives one week after OpenAI shut down Sora, as the AI content creation market shifts. Music generation is now a race between Google, Suno, and ElevenLabs, with Udio pivoting to a licensed remix model after settling with Universal Music Group.

Why It Matters for Creators

Three minutes changes the use case entirely. Thirty-second clips work for social media intros and sound effects, but three minutes covers a full YouTube background track, podcast segment, or short film score. The structural controls mean you can prompt for a build-up in the intro and a drop in the chorus rather than hoping the AI handles pacing on its own.

Developer API access is the bigger signal. Apps like Suno V5 Studio built their own models, but smaller creative tools can now plug into Google's music generation without training anything. Video editors, game engines, and podcast platforms could add "generate a soundtrack" as a native feature through the Gemini API.

What to Do Next

Paid Gemini subscribers can try Lyria 3 Pro now through the Gemini app. Developers can access Lyria 3 through the Gemini API paid preview or test it free in Google AI Studio. If you build creative tools, the API opens a new category of music-generation-as-a-service that previously required building or licensing a dedicated model.

This story was covered by Creative AI News.

Subscribe for free to get the weekly digest every Tuesday.