Google integrates lyria 3.5 music AI into gemini for full song generation

On September 5, 2026, Google announced the integration of its latest music generation model, Lyria 3.5, into the Gemini application and API, marking a shift from internal use to broader developer and user access. The update allows developers to leverage Lyria 3.5 through Google AI Studio, Vertex AI, and the Gemini API. The model can generate full songs lasting several minutes, including verses and choruses, while Lyria 3 Clip produces 30-second loops for preview. Both models default to MP3 output, with Lyria 3.5 also supporting WAV. Developers can input text prompts alongside up to 10 images to guide music creation. All generated content is managed via the Interactions API and includes SynthID watermarks for copyright tracking. Google emphasized the importance of content authenticity, ensuring the watermarks resist common audio processing techniques. The release highlights Google's expansion into mainstream music AI, offering tools for both creators and developers to explore new possibilities in automated music composition.