MiniMax Music·by MiniMaxMusicNew
MiniMax Music 3.0
MiniMax Music 3.0 — a finished, mixed song of up to 5 minutes in one generation: vocals, harmonies and full backing. Write the style as the prompt; put your words in the Lyrics field, or leave it empty and the model writes them from the prompt. Section tags like [Verse] / [Chorus] are supported.
Parameters
| Parameter | Type | Default | Range or options |
|---|---|---|---|
Bitrate (mp3) bitrateOutput bitrate (bps). Applies to mp3 only. | select | 256000 | 32000, 64000, 128000, 256000 |
Format formatOutput audio format. | select | mp3 | mp3, wav |
Instrumental only is_instrumentalGenerate instrumental music only (no vocals). Lyrics are ignored. | select | false | true, false |
Lyrics lyricsLeave empty and the model writes the lyrics from your prompt. Section tags such as [Verse] or [Chorus] on their own line shape the arrangement. Up to 3,500 characters. | textarea | — | — |
Sample rate sample_rateOutput sample rate (Hz). | select | 44100 | 16000, 24000, 32000, 44100 |
Inputs
No reference files: this model works from the prompt alone.