Google has introduced Lyria 3 Pro, an advanced version of its AI music generation model, on March 25, 2026. The new model enables users to create tracks up to three minutes long with enhanced structural awareness and customization. Lyria 3 Pro can generate specific musical elements such as intros, verses, choruses, and bridges, offering creators greater control and flexibility. The rollout extends Lyria’s capabilities to more Google products, including Vertex AI, Google AI Studio, Gemini API, Google Vids, the Gemini app, and ProducerAI.

According to Google, Lyria 3 Pro is designed to make high-quality music generation accessible to developers, businesses, and creators. It is available in public preview on Vertex AI for scalable audio production and integrated into Google Vids for Workspace and AI Pro subscribers. ProducerAI users can now collaborate with the model to iterate on full-length songs. The company emphasized responsible development, using only materials it has rights to and embedding all outputs with SynthID watermarks to identify AI-generated content.

Google stated that Lyria 3 Pro was developed in collaboration with musicians through its Music AI Sandbox, ensuring the technology supports creative expression while protecting intellectual property rights.