For coding agents, discover the current recommended music shortlist first with
GET /v1/models?recommended_for=music, then send the selected model explicitly to this endpoint.poll_url for status checks.
Request Body
Model to use:
suno_music for music generation, suno_lyrics for lyrics only.Music description. Required for music, lyrics, upload-cover, and upload-extend requests; omit for add-instrumental.
Title for the generated song.
Style tags (e.g., “pop, upbeat, electronic”).
Styles to avoid. Required when
audio_operation is add-instrumental.Generation type:
MUSIC (default) or LYRICS.Official Suno model version. Required for music generation (when
action is omitted or MUSIC); omit for lyrics-only requests. Supported values: chirp-v3-5, chirp-v4, chirp-v4-5, chirp-v4.5, chirp-v4-5plus, chirp-v4.5plus, chirp-v4-5all, chirp-v4.5all, chirp-v5, chirp-v5-5, chirp-v5.5.Set to
true to generate instrumental output without vocals.Publicly reachable reference or uploaded audio URL. When provided without
audio_operation, TokenLab uses the upload-cover flow.Uploaded-audio mode:
upload-cover, upload-extend, or add-instrumental. upload-extend requires continue_at; add-instrumental requires audio_url, title, tags, and negative_tags.ID of a previous clip to continue from.
Timestamp (in seconds) to continue from. Required when
audio_operation is upload-extend.Response
Task ID for polling status.
Async task identifier alias when returned by the adapter.
Preferred polling URL for this task when provided.
Task status:
pending, processing, completed, or failed.Unix timestamp of task creation.