On September 5, 2026, Google officially launched the new generation music generation model Lyria 3.5 in Gemini applications and APIs. This model is available to global users and developers and can be accessed through Google AI Studio, Vertex AI, and Gemini API. Lyria 3.5 supports the generation of complete songs lasting several minutes, as well as 30-second clip versions called Lyria 3 Clip. The default output format is MP3, but WAV format is also supported. In addition to text prompts, users can upload up to 10 images to guide music creation. All generated audio is scheduled through the Interactions API and includes anti-compression SynthID hidden watermarks for copyright tracing.
On September 5, 2026, Google officially launched the music generation model Lyria 3.5 in Gemini applications and APIs. This model is available to global users and developers and can be invoked through Google AI Studio, Vertex AI, and Gemini API. Lyria 3.5 supports generating complete songs lasting several minutes, as well as 30-second clip versions of Lyria 3 Clip. The default output format is MP3, but WAV format is also supported. Users can upload up to 10 images to guide music creation in addition to text prompts. All audio is scheduled using the Interactions API, and anti-compression SynthID hidden watermarks are used for copyright tracing.