Lyria 3 by Google Deepmind

Generate 30-second songs from text or images in Gemini

Visit Lyria 3 by Google Deepmind →

Lyria 3 is Google DeepMind's music generation model integrated into the Gemini app, allowing users to create 30-second songs with vocals and cover art from text prompts or photos. It is currently in beta and accessible via the Gemini app or website. It is notable for being one of the most advanced AI music generation models, combining multimodal input with full song output including vocals.

At a glance

Company
Google DeepMind
Pricing
freemium
API available
No
Self-hostable
No
Launched
2026-02
Last verified
2026-05-11

Capabilities

text-to-musicimage-to-musicvocal-generationcover-art-generationmultimodal-input

Categories

audio

Alternatives

For AI agents: machine-readable markdown version of this page at /tools/lyria-3-by-google-deepmind.md, or send Accept: text/markdown.