All models
Music generationPreview
Google DeepMindLyria 3.5
Google's flagship music model for complete, high-fidelity songs with richer arrangements, expressive vocals, structured lyrics, and detailed creative control.
Plain-English overview
What Lyria 3.5 actually is
Lyria 3.5 turns a written music brief or as many as ten images into a finished stereo track. It can compose instrumentals or songs with vocals, generate lyrics or follow lyrics you provide, and organize a piece into intros, verses, choruses, bridges, and outros.
The Gemini API returns 44.1 kHz audio as MP3 or WAV alongside lyrics and song-structure text. Prompts can specify genre, instruments, tempo, musical key, mood, duration, and timestamped changes, making the model useful for both quick ideas and more directed production briefs.
Good fit for
- Brand jingles and video backing tracks
- Complete vocal songs with custom structure
- Turning visual concepts into music
Category comparison
The facts that matter for music models
These are provider-published specifications, not Cody benchmark scores. Follow the linked sources for current limits and endpoint-specific exceptions.
- Generation price
- $0.08 / full songCurrent provider price for one song or clip, including the listed duration basis.
- Song length
- Up to 3 minutes in Google apps; API is prompt-controlledMaximum or typical generated-track duration documented by the provider.
- Audio output
- 44.1 kHz stereo · MP3 or WAVPublished sample rate, channel format, and downloadable file types.
- Vocals & lyrics
- Vocals or instrumental; generated/custom lyricsWhether the model creates vocals, instrumentals, generated lyrics, or user-supplied lyrics.
- Creative control
- Text, up to 10 images, BPM, key, structure, timestampsPrompt, image, tempo, key, structure, timeline, reference, or editing controls documented for the model.
- Where to use it
- Gemini Interactions API, AI Studio, Gemini, Flow MusicDirect API and notable first-party creative products included in this guide.
API and provider access
Where to get Lyria 3.5
Gemini APILyria 3.5 API documentationOfficial preview model ID for the Interactions API.Google AI StudioTry Lyria 3.5 in AI StudioInteractive developer access using the Gemini API project and billing model.Gemini and Flow MusicExplore Lyria 3.5 productsConsumer and creator experiences; features and plan limits can differ from the API.
Availability
Regions and access stage
Available through the Gemini API and Google AI Studio, and in the Gemini app, Google Flow Music, and Google Vids.
Google describes Gemini app access as global wherever Gemini is available; API access follows the live Gemini API region list and account eligibility.
Check live availabilityData and training
The route matters.
Lyria 3.5 has no free API tier. Google marks paid Gemini API content as not used to improve its products; consumer-product data terms may differ from the paid developer API.
This is a concise reading of the cited provider material, not legal advice. A third-party gateway can have different storage, routing, training, and residency terms from the model maker's direct API.
Read the provider policyFrequently asked questions
Lyria 3.5 FAQ
What is Lyria 3.5?
Google's flagship music model for complete, high-fidelity songs with richer arrangements, expressive vocals, structured lyrics, and detailed creative control. Lyria 3.5 turns a written music brief or as many as ten images into a finished stereo track. It can compose instrumentals or songs with vocals, generate lyrics or follow lyrics you provide, and organize a piece into intros, verses, choruses, bridges, and outros.
When was Lyria 3.5 released?
Lyria 3.5 was released on July 29, 2026 according to the cited provider materials.
Where can I access Lyria 3.5?
Available through the Gemini API and Google AI Studio, and in the Gemini app, Google Flow Music, and Google Vids. The access routes listed in this guide are Gemini API, Google AI Studio, and Gemini and Flow Music.
How much does Lyria 3.5 cost?
$0.08 / full song. Google lists Lyria 3.5 at $0.08 per paid API request with no free API tier. Each request generates one full song; pricing for Gemini, Flow Music, or Google Vids follows those products' plans.
Where is Lyria 3.5 available?
Available through the Gemini API and Google AI Studio, and in the Gemini app, Google Flow Music, and Google Vids. Google describes Gemini app access as global wherever Gemini is available; API access follows the live Gemini API region list and account eligibility.
Is my Lyria 3.5 API data used for training?
Lyria 3.5 has no free API tier. Google marks paid Gemini API content as not used to improve its products; consumer-product data terms may differ from the paid developer API. The policy belongs to the provider route and account terms, so verify it again before production use.
Related comparisons
Head-to-head comparisons
Lyria 3.5 vs Eleven Music v2
ElevenLabs' current studio-grade music model for complete vocal or instrumental tracks, with structured composition, section editing, audio references, and a direct API.
Open full comparisonLyria 3.5 vs Seed-Music
ByteDance Seed's research music suite for expressive multilingual vocals, multimodal creative direction, audio continuation, style transfer, and fine-grained editing.
Open full comparisonLyria 3.5 vs HappyShrimp 1.0
Alibaba's web-based music beta for turning a detailed prompt or lyrics into complete vocal or instrumental songs.
Open full comparisonLyria 3.5 vs Suno v5.5
Suno's latest personalized song model for richer arrangements and vocals, with verified Voices, custom models, taste controls, and a full music editor.
Open full comparison
