Skip to main content
POST
Generate music

Authorizations

x-api-key
string
header
required

Your Recoup API key. Learn more.

Body

application/json
prompt
string
required

Music description: style, mood, vocals, instrumentation and arrangement.

Example:

"Genre: acoustic pop. BPM: 96. Key: C major. Warm and intimate, building gently into the chorus."

lyrics
string
required

The lyrics to sing. Structure tags such as [intro], [verse], [chorus] and [outro] must each be on their own line; text on the same line as a leading tag is dropped by the model.

Example:

"[verse]\nMorning light filtering through the pine\n[chorus]\nSoftly the world begins to breathe"

duration
number
default:60

Upper bound on the generated audio length in seconds. The model may stop earlier; the actual length comes back as duration_seconds.

Required range: 10 <= x <= 300
seed
integer

Seed for reproducibility. Omit for a random seed.

num_inference_steps
integer
default:30

Flow-matching steps per denoising chunk. More steps improve quality at the cost of speed.

Required range: 1 <= x <= 100
guidance_scale
number
default:1.7

Classifier-free guidance scale of the flow-matching stage.

Required range: 0 <= x <= 20
account_id
string<uuid>

Optional. Generate on behalf of another account you can access (self, shared organization, or admin). Defaults to the calling account. Organizations are accounts, so this is also how an organization generation is scoped: send the organization's account id.

Response

Generation accepted and queued.

The accepted generation. Poll it until status is terminal.

status
enum<string>

Status of the request

Available options:
success
generation
object

One music generation.