Lyria 3.5 API: model id, request and how to call it
Lyria 3.5 is in the Gemini API as lyria-3.5. Here is what Google's page says about it and how to call it from Sume with one music request.

The Lyria 3.5 API is Google's music model, offered in the Gemini API under the model id lyria-3.5. It writes a full-length song from a text prompt, optionally with images. On Sume you reach it with one request to POST /v1/music-router/generate: leave model out (or send sume/music-auto, which resolves to Lyria 3.5 today) or pin lyria-3.5.
Google's side comes from its Lyria page and announcement; Sume's from the Music Router and Music 1.0 docs. All read 2026-09-29.
What does Google's Lyria 3.5 API page say?
| Topic | What Google says |
|---|---|
| Model id | lyria-3.5 |
| Output | Full-length songs, "a couple of minutes", MP3 by default |
| Length | Controlled through the prompt |
| Inputs | Text, with up to 10 images |
| Structure | Section tags such as [Verse], [Chorus], [Bridge] |
| Watermark | All generated audio carries a SynthID watermark |
| Editing | Single-turn; iterative editing of a result is not supported |
| Released in Gemini | 2026-09-04; the announcement lists the Gemini app, Google Flow Music, AI Studio and Google Vids |
How do I call Lyria 3.5 through Sume?
Send a prompt. The body has no seed, temperature or duration field; the docs say to steer length in the prompt and reject duration and duration_seconds.
curl -X POST https://api.sume.com/v1/music-router/generate \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: lyria-35-api-001" \
-d '{
"model": "lyria-3.5",
"prompt": "Warm lo-fi hip hop, 84 BPM, C minor. Dusty Rhodes chords, brushed boom-bap drums. A 30-second track. Instrumental, no vocals."
}'How do I get the audio back?
The call returns a job. Poll GET /v1/jobs/{id}/status, then read GET /v1/jobs/{id}/result. The audio is the entry in result.artifacts[] whose type is audio, typically audio/mpeg on media.sume.com. result.lyrics carries the model-reported lyrics or section map when present. job.request.routed_model names the engine that ran, so you can confirm it was lyria-3.5.
Which request fields exist?
prompt: required, 1-5000 characters. Put exclusions in the positive prompt.image_url: optional public HTTPS image for visual conditioning.negative_prompt: not supported when non-empty (HTTP 400).mode,webhook_url,wait_timeout_seconds: how you learn the result.
Is Sume's price the same as Google's?
Google's own prices are on its Gemini API pricing page; Sume bills every Music Router model at its fixed music price per generation, listed on API pricing. The two are separate price lists, so check both before you plan volume.
Sources
Related posts
More in Models
- AI music generator with vocals: Lyria 3.5 lyrics by API
Lyria 3.5 can sing. How to ask for vocals or an instrumental, steer lyrics with section tags, and read the lyrics back from a Sume music job.
- Lyria 3.5 release date: where it is available
Google announced Lyria 3.5 in the Gemini app on 2026-09-04. Where Google says you can use it, and where Sume lists it.
- How long is a Lyria 3.5 song, and how do I set the length?
Lyria 3.5 makes songs of a couple of minutes, and length is set in the prompt. What Google says, what Sume rejects, and prompts that steer duration.
- 2K AI video generator API: MiniMax H3 at 2K on Sume
MiniMax H3 makes video up to 2K. On Sume you ask for resolution 2K on minimax-h3, an upscale of native 768p. Request, price for 10 seconds, and limits.
Written by Sume