| 提示词 | string | A description of the desired audio content. The prompt will be strictly used as the lyrics and sung in the generated track. Max 5000 characters for V6 models. | Hard-hitting rap track with aggressive beat and confident male vocals about winning. |
| 风格 | string | Music style specification for the 生成的音频. | Classical |
| 纯音乐 | boolean | 启用此选项可在没有提示词的情况下生成音乐。如果为 false,提示词会作为确切歌词使用。 | true |
| 模型 | 枚举(10 个选项) | The AI model version to use for generation. V6 offers enhanced musical expression and natural vocals, V6_MINI is a lightweight/fast variant, and V6_WILD pushes creative boundaries for bolder expression. | V6 |
| 时长 | int | Track duration (秒) (optional). Range 10–360. Supported in custom mode with model V5_5, V6, V6_MINI, or V6_WILD. Leave unset to let the model choose the duration. | 20 |
| 人物 ID | string | 要应用于生成音乐的 Persona ID 或自定义声音 ID(可选)。请与 persona_model 搭配使用以消除歧义。 | - |
| Persona 模型 | 枚举(2 个选项) | What kind of persona_id this is. Set to voice_persona when persona_id is a cloned voice ID from suno-voice-clone. Requires model V5, V5_5, V6, V6_MINI, or V6_WILD. | - |
| 参考音频 URL | array | Reference audio track(s) whose melody, rhythm, and mood should inspire the generated music (optional). Duration 6s–30min, up to 500MB per file. Only used in Non-custom Mode (custom_mode set to false); combined with video_urls, must not exceed 10 total attachments. | undefined |
| 参考视频 URL | array | Reference video whose atmosphere and story should inspire the generated music (optional). MP4/MOV/WEBM, up to 1 file, max 100MB, max 241s. Only used in Non-custom Mode (custom_mode set to false); combined with audio_urls, must not exceed 10 total attachments. | undefined |