Google DeepMind has brought its latest music generation model, Lyria 3, into the Gemini app. The company describes Lyria 3 as its most advanced music model to date. With it, users can generate 30-second tracks by providing text or images as prompts.
DeepMind frames the feature as a way to let anyone express themselves musically, lowering the barrier to creating short pieces of music. The integration means music generation is now part of the consumer-facing Gemini experience rather than a separate research demo.
Because this is a single announcement, there are no independent comparisons or contrasting details to weigh. The only specifics provided are the model name, the 30-second length, the text/image input modes, and the claim that it is DeepMind's most advanced music generation model.