Skip to main content
POST
Flow Music
flow-music generates one track with Flow Music, which runs Google Lyria. Describe the sound in sound_prompt, supply lyrics, or both, and optionally set the tempo and length. Each request returns exactly one clip, which later Flow Music operations can extend, rework, split, export, or render as a video. Confirm live rates on the SnapGen models page. Failed tasks are not charged.
string
required
Set to flow-music.
string
default:"lyria-3-pro"
Lyria version: lyria-3-pro (Lyria 3 Pro) or lyria-3.5 (Lyria 3.5).
string
Title of the new clip, up to 120 characters.
string
Style or sound description, up to 2,000 characters, for example upbeat pop music with piano. Send sound_prompt, lyrics, or both.
string
Lyrics, up to 3,000 characters. flow-music-lyrics can write them.
integer
Tempo in beats per minute, from 1 to 300.
integer
Length of the track in seconds, from 1 to 240.
integer
Random seed, from 0 to 4,294,967,295. The same seed and request give similar, not identical, results.
string
Optional public https URL that receives one signed event when the task finishes. See Task webhooks.
string
Required with webhook: 16 to 256 characters, used to sign the delivery.

Request rules

The gateway checks these before it reserves any balance:
  • Send sound_prompt, lyrics, or both.

Request

To sing your own words with Lyria 3.5, add lyrics and select the version:
The response is HTTP 202 with a task object and a Location: /v1/tasks/{id} header. Poll GET /v1/tasks/{id} every 10 to 20 seconds until status is succeeded or failed.

Result

One clip in result.tracks[0], with its audio_url and, when the provider returns them, wav_url and duration_seconds. result.urls lists the audio file.

Next steps

Pass this task’s id as source_task_id to another Flow Music model. Each edit returns a new clip, so chain the latest task for the next step.
Result URLs are kept for about 72 hours. Download the files you want to keep. See Music generation tasks for task references, webhooks, and errors.