Fordel StudiosBuild with Lyria 3, our newest music generation model
What Happened
Lyria 3 is now available in paid preview through the Gemini API and for testing in Google AI Studio.
Our Take
Google released Lyria 3 — a music generation model — into paid preview via the Gemini API, with free testing available in AI Studio. It generates full audio tracks from text prompts.
If you're running a video generation pipeline using Runway or Sora, you're still paying per-track licensing through Epidemic Sound or similar at $150–300/month. Lyria 3 changes that math. Most teams default to stock licensing out of habit, not necessity — that habit now has a dollar cost you can calculate.
AI video pipeline builders should benchmark Lyria 3 against current licensing bills this week. Pure text or code tools have no use case here.
What To Do
Swap Epidemic Sound licensing for Lyria 3 via the Gemini API in AI video pipelines because per-generation cost will undercut flat subscription fees at any meaningful output volume.
Builder's Brief
What Skeptics Say
'Paid preview' access limits real adoption signal — Google controls who gets in, so early enthusiasm is curated, not organic. API economics for music gen remain unproven at the volumes needed for sustainable B2C apps.
Cited By
React
Get the weekly AI digest
The stories that matter, with a builder's perspective. Every Thursday.