Lyria 3

Overview

Lyria 3 is a generative music and audio model from Google that creates original music from text prompts and, in some cases, images. Across the newsletter mentions, it appears first as a launch tied to Google AI and Google DeepMind, then as a beta capability in the Gemini app, and later as a developer-accessible model exposed via API. The model is described as producing dynamic music, 30-second songs, and longer full-length tracks in product surfaces such as Gemini.

For AI Product Managers, Lyria 3 matters because it shows how generative media is moving from consumer-facing demos into deployable platform capability. It is relevant not just as a creative tool, but as infrastructure for building music generation, soundtrack creation, multimodal experiences, and branded audio workflows into apps and products. The addition of SynthID watermarking and API availability also makes it noteworthy from both product strategy and trust-and-safety perspectives.

Key Developments

• 2026-02-19: Google DeepMind rolled out Lyria 3 in beta globally via the Gemini app for custom audio generation. Outputs included SynthID, Google's invisible watermark for AI-generated content.
• 2026-02-21: Google AI highlighted Lyria 3 alongside Gemini 3.1 Pro and other launches, describing it as a model that turns photos or text into dynamic music.
• 2026-04-10: Josh Woodward announced full access to Lyria 3 in the Gemini app, with up to 5 free full-length tracks per day, unlimited 30-second clips, and paid upgrades for more music and related media generation.
• 2026-04-11: DeepLearning.AI described Lyria 3 as an AI music generator that transforms text prompts or images into original 30-second songs.
• 2026-05-26: Logan Kilpatrick announced that Lyria 3 is available via API, enabling developers to build directly with the model.

Relevance to AI PMs

• Prototype multimodal media features: Lyria 3 is useful for testing user experiences where text or images generate music, such as soundtrack creation, social content tools, creative assistants, game prototyping, or marketing asset workflows.
• Plan product packaging and monetization: The Gemini rollout signals concrete packaging choices—free daily quotas, short-form clips, and paid upgrades—which AI PMs can study when designing usage limits, conversion paths, and premium tiers for generative media products.
• Address trust, attribution, and governance early: The mention of SynthID watermarking makes Lyria 3 relevant for PMs working on provenance, disclosure, rights-sensitive content workflows, and internal policies for AI-generated media.

Related

• Google / Google AI / Google DeepMind: The organizations associated with launching and distributing Lyria 3 across announcements and product channels.
• Gemini app / GeminiApp / gemini-app: Early distribution surface where users could access Lyria 3 in beta and later with broader usage allowances.
• API: Marks the transition from end-user feature to developer platform capability.
• SynthID: Google's invisible watermarking system applied to Lyria 3 outputs, relevant for authenticity and responsible AI considerations.
• Josh Woodward: Shared product rollout details for Gemini app access and usage tiers.
• Logan Kilpatrick: Announced API availability for developers.
• DeepLearning.AI: Helped amplify the launch framing of Lyria 3 as an AI music generator.