Google Gemini Adds Lyria 3.5 Music AI

Google has expanded its Gemini artificial intelligence platform with the integration of Lyria 3.5, its most advanced music generation model, allowing users to create full-length songs and soundtracks from text or image prompts. The rollout, announced on September 4, 2026, makes the technology available to all Gemini users globally through the web, mobile app, and the Gemini API, following an initial launch in Google Flow Music on July 29, 2026. The new model represents a significant upgrade from its predecessor, Lyria 3, which was limited to generating 30-second clips. Lyria 3.5 can produce complete musical works at 44.1 kHz stereo quality, featuring verses, choruses, and bridges. Google states the model offers richer arrangements, more expressive and realistic vocals, and a better understanding of user prompts. Within the Gemini app, users can select a genre, describe a desired style, choose between vocal or instrumental approaches, and specify whether they want a short clip or a longer track. Templates are also available to help beginners generate background music for videos, custom ringtones, or other themes.