AI News: The AI World is REALLY Scared Right Now
Google's new music model, available in Gemini, Flow Music, AI Studio and Google Vids. The creator played a sample.
Lyria 3.5 at a glance
- What is Lyria 3.5?
- Lyria 3.5 is a model from Google. Google DeepMind's music generation model producing vocal tracks up to 3 minutes from text or image prompts.
- Who makes Lyria 3.5?
- Lyria 3.5 is made by Google.
- Does Lyria 3.5 work in the EU?
- Unverified. No explicit evidence found; the provided pages contain no supported-countries list or EU restriction for Lyria 3.5. Last verified 14 Nov 2026.
- How much does Lyria 3.5 cost?
- Pricing unknown. No pricing in the provided material; access is via Google products (Gemini, Flow Music, AI Studio, Vids). Licence: Proprietary.
- Where was Lyria 3.5 shown?
- In 1 creator video by Matt Wolfe. Most recently by Matt Wolfe on 11 Sept 2026, at 31:34 in “AI News: The AI World is REALLY Scared Right Now”.
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
About
Lyria 3.5 is Google DeepMind's latest music generation model. It generates full tracks up to three minutes long from text prompts, with improvements over earlier versions in musicality, lyrics and vocal quality. Users can specify genre, tempo, key, instrumentation and vocal style, and can choose a target duration (for example a 60-second clip or a full 3-minute song).
The model supports vocals in multiple languages, user-written or generated lyrics, and image-to-music generation, where an uploaded image is turned into a track. Lyria belongs to a model family that also includes Lyria RealTime for interactive, streaming generation and Magenta RealTime, an open model.
Per creator coverage, Lyria 3.5 is available in Gemini, Flow Music, Google AI Studio and Google Vids. All generated tracks carry SynthID watermarks so AI-created or edited music can be detected. Google notes that the model has limitations and that users should check outputs against their intent. Pricing and regional availability details were not given in the provided material.
Key features
- Tracks up to 3 minutes with selectable duration
- Text-to-music with control over genre, BPM, key and vocal style
- Image-to-music generation
- Multilingual vocals and lyrics generation or custom lyrics
- SynthID watermark on all outputs
- Available in Gemini, Flow Music, AI Studio and Google Vids
1 moment
As seen on YouTube
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
Keep exploring
Similar AI tools
Closest neighbours by meaning, not just by category.
EM
ElevenLabs Music v2.5
Model · ElevenLabs
ElevenLabs' text-to-music model that generates tracks with or without vocals from natural-language prompts.
Available in EUMusic
Release date n/aFreemium
MM
MiniMax Music 3
Model · MiniMax
Open-weights text-to-music model from MiniMax that generates songs with vocals and lyrics across genres and languages.
Available in EUMusic
Release date n/aOpen source
AC
ACE-Step
Model
Open-source diffusion-based music generation foundation model with lyric-aligned vocals and editing controls.
Available in EUMusic
6 May 2025Open source
EM
ElevenLabs Music v2
Model · ElevenLabs
ElevenLabs' text-to-music model for generating songs with vocals, inpainting and genre transitions, cleared for commercial use.
Available in EUMusic
Release date n/aFreemium
KH
Khala
Model
Open-source text-to-music model that generates full songs with vocals from a text prompt and lyrics.
UnverifiedMusic
Release date n/aOpen source
A1
ACE-Step 1.5 XL
Model · ACE Studio
Open-source local music generation model with a 4B-parameter DiT decoder, extending ACE-Step 1.5 with higher audio quality.
Available in EUMusic
2 Apr 2026Open source