Music Generation

The Suno music generation endpoint. Submit a song generation task via POST /suno/submit/MUSIC (custom lyrics mode or description mode). The interface is async: on success it returns a task_id (prefixed with task_); poll the Query Result endpoint with that ID until status becomes SUCCESS to fetch the final audio URL.

Billing model

This endpoint is billed per submitted action under the model name suno_music, deducted from your account balance. Add the models first on the channel (select channel type TTAPI and add suno_music and suno_lyrics).

Two generation modes

The request body has no model field — the mode is chosen by the parameters:

Custom lyrics mode: fill in prompt (the lyrics), optionally title and tags (style tags). Providing prompt enables this mode;
Description mode: omit prompt and provide gpt_description_prompt instead (a description such as "night city lo-fi piano"), letting the upstream generate lyrics and compose the song automatically.

Other common parameters: make_instrumental (instrumental-only track) and mv (model version).

mv model versions

Supported values: chirp-v5-5, chirp-v5, chirp-v4-5+, chirp-v4-5, chirp-v4-5-all, chirp-v4, chirp-v3-5, chirp-v3-0. Defaults to chirp-v5; you can also map suno_music to a fixed version in the channel's model mapping.

Extend an existing song

MUSIC only creates new songs and does not accept task_id, continue_clip_id, or continue_at. Use EXTEND or UPLOAD_EXTEND on the Edit and Extend page instead.

Authorization

BearerAuth

AuthorizationBearer <token>

使用 Bearer Token 认证。格式: Authorization: Bearer sk-xxxxxx 令牌需开通对应模型的调用权限。

In: header

Request Body

application/json

prompt?string

歌词正文（自定义歌词模式）。提供后自动进入自定义歌词模式

gpt_description_prompt?string

灵感描述（灵感描述模式）。不提供 prompt 时必填，例如「城市夜景 Lo-Fi 钢琴」

title?string

歌曲标题（自定义歌词模式，最长 80 字符）

tags?string

风格标签，逗号分隔（自定义歌词模式）

make_instrumental?boolean

true 生成纯音乐（不带人声）

custom?boolean

显式指定自定义歌词模式；纯器乐请求可与 instrumental=true 一起使用

instrumental?boolean

是否生成纯音乐；优先于兼容字段 make_instrumental

mv?string

模型版本。可选：chirp-v5-5、chirp-v5、chirp-v4-5+、chirp-v4-5、chirp-v4-5-all、chirp-v4、chirp-v3-5、chirp-v3-0。默认 chirp-v5

negative_tags?string

不希望出现的风格标签

style_weight?number

Range0 <= value <= 1

weirdness_constraint?number

Range0 <= value <= 1

audio_weight?number

Range0 <= value <= 1

duration?integer

Range10 <= value <= 360

auto_lyrics?boolean

vocal_gender?string

Value in"Male" | "Female"

persona_id?string

audio_format?string

Value in"mp3" | "m4a" | "wav"

hookUrl?string

任务完成回调地址

isStorage?booleanDeprecated

Response Body

application/json

curl -X POST "https://www.omwai.xyz/suno/submit/MUSIC" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "[Verse]\nNeon headlights on a rainy street\n[Chorus]\nWe are the midnight drive",
    "title": "Midnight Drive",
    "tags": "synthwave, female vocal, cinematic",
    "mv": "chirp-v5"
  }'

curl -X POST "https://www.omwai.xyz/suno/submit/MUSIC" \  -H "Content-Type: application/json" \  -d '{    "prompt": "[Verse]\nNeon headlights on a rainy street\n[Chorus]\nWe are the midnight drive",    "title": "Midnight Drive",    "tags": "synthwave, female vocal, cinematic",    "mv": "chirp-v5"  }'

{
  "code": "success",
  "message": "success",
  "data": "task_01J8ABCDEFGHIJKLMNOPQRST"
}

{
  "code": "invalid_request",
  "message": "string"
}

Async flow

POST /suno/submit/MUSIC submits and returns code=success with data (task_id, prefixed with task_);
Poll GET /suno/fetch/{task_id} with that task_id;
When status=SUCCESS, result_url holds the final audio URL and data contains the song details; when FAILURE, fail_reason explains why.

Music Generation

Two generation modes

Authorization

Request Body

Response Body

200application/json

400application/json

application/json

application/json