Google DeepMind and Google announce the launch of Lyria 3.5, a new music generation model, in Google Flow Music. The announcement says Lyria 3.5 introduces improvements aimed at producing more musical results, with upgraded lyric handling, higher-quality vocals, and expanded creative control for users. Across the sources, the model is presented as the newest version in the Lyria series and is positioned as enabling users to craft “richer” or more developed tracks compared with earlier capabilities. While the outlets emphasize different aspects—such as musicality, lyrics, vocal quality, and creative control—they align on the core message: Lyria 3.5 is being integrated into Google Flow Music as the latest model for generating music. The announcements describe the release as a product update for the platform rather than a standalone research release, and they focus on user-facing performance areas rather than detailed technical specifications. No additional context on availability timelines, supported languages, or evaluation metrics is provided in the excerpts.