Lyria 3.5 is Google DeepMind's new text-to-music model. It writes complete songs — melody, instrumentation, and vocals — from a text prompt, and it rolled out on July 29, 2026 inside Google Flow Music, DeepMind's free creative studio for music.

Where can I try Lyria 3.5?

Lyria 3.5 is available inside Google Flow Music at flowmusic.google, Google's free music creation product. It is not offered as a standalone API in the announcement; you use it through Flow Music's prompt interface to write, refine, and export tracks.

Is Lyria 3.5 free to use?

Yes. The announcement says Lyria 3.5 is rolling out in Google Flow Music, which is a free product. Google's wider Flow creative suite has paid tiers for video and image work, but the Lyria 3.5 music launch is described as accessible via flowmusic.google at no cost.

What is new in Lyria 3.5 compared to earlier Lyria models?

Google names four upgrades for Lyria 3.5: richer melodic structure, better lyric writing with structural awareness, more expressive vocals, and finer creative control over tempo and duration. It sits alongside Lyria RealTime and Magenta RealTime in DeepMind's Lyria family.

Can Lyria 3.5 generate vocals?

Yes. Google's announcement highlights vocals as one of the four areas Lyria 3.5 improves, calling out more expression, emotion, realism, and pronunciation accuracy. DeepMind's Lyria technology page also notes control over vocal style across genres and languages.

Google DeepMind · 2026-07-29 · major

Lyria 3.5 — Google DeepMind's new music model in free Flow Music

Google DeepMind announced Lyria 3.5, its new text-to-music model, and started rolling it out in Google Flow Music. Google says the upgrade improves melodic structure, lyric writing, vocal expression, and creative control over tempo and length.

Google DeepMind Lyria 3.5 announcement card, Google Flow Music branding

Google DeepMind's new music model — richer melodies, clearer vocals, longer songs in the free Flow Music app.

Quick facts

Maker	Google DeepMind
Version	Lyria 3.5
Available in	Google Flow Music (flowmusic.google)
Price	Free
Song length	Up to ~3 minutes
Family	Lyria, Lyria RealTime, Magenta RealTime
Announced	July 29, 2026

What is it?

Lyria 3.5 is Google DeepMind's newest text-to-music model. It writes full songs from a prompt — melody, backing, and vocals — inside Google Flow Music, the free creative studio at flowmusic.google. It joins the wider Lyria family that also includes Lyria RealTime and Magenta RealTime.

How does it work?

The 3.5 update focuses on four things Google names in the announcement: richer melodic structure, better lyric writing with structural awareness, more expressive and emotional vocals, and finer control over tempo and duration. DeepMind's Lyria technology page describes the family as producing tracks up to about three minutes long across multiple genres.

Why does it matter?

For hobby creators, Lyria 3.5 raises the ceiling on what a free music tool can generate — from short loops toward full-length arrangements with expressive vocals. It sharpens Google's answer to Suno and Udio and pushes DeepMind's audio research closer to a mass-market product.

Who is it for?

musicians, creators, hobbyists

Frequently asked questions

What is Lyria 3.5?: Lyria 3.5 is Google DeepMind's new text-to-music model. It writes complete songs — melody, instrumentation, and vocals — from a text prompt, and it rolled out on July 29, 2026 inside Google Flow Music, DeepMind's free creative studio for music.
Where can I try Lyria 3.5?: Lyria 3.5 is available inside Google Flow Music at flowmusic.google, Google's free music creation product. It is not offered as a standalone API in the announcement; you use it through Flow Music's prompt interface to write, refine, and export tracks.
Is Lyria 3.5 free to use?: Yes. The announcement says Lyria 3.5 is rolling out in Google Flow Music, which is a free product. Google's wider Flow creative suite has paid tiers for video and image work, but the Lyria 3.5 music launch is described as accessible via flowmusic.google at no cost.
What is new in Lyria 3.5 compared to earlier Lyria models?: Google names four upgrades for Lyria 3.5: richer melodic structure, better lyric writing with structural awareness, more expressive vocals, and finer creative control over tempo and duration. It sits alongside Lyria RealTime and Magenta RealTime in DeepMind's Lyria family.
Can Lyria 3.5 generate vocals?: Yes. Google's announcement highlights vocals as one of the four areas Lyria 3.5 improves, calling out more expression, emotion, realism, and pronunciation accuracy. DeepMind's Lyria technology page also notes control over vocal style across genres and languages.

Try it

flowmusic.google