ModelRadar
/

Music3

MiniMaxπŸ‡¨πŸ‡³ CNMajor labΒ· Aug 7, 2026

Audioopen weights

MiniMax Music 3 generates complete songs up to five minutes long from lyrics and a style description, using an 8B global LLM for song structure and a 0.6B local LLM for acoustic detail.

Summary of the model card

Modalities
text β†’ audio

1 variant

StandardHF

Sources

HuggingFaceBlog
Deutsch

MiniMax Music 3 erzeugt vollstaendige Songs mit bis zu fuenf Minuten Laenge aus Liedtext und Stilbeschreibung. Ein 8B Global LLM steuert die Struktur, ein 0,6B Local LLM die akustischen Details, die Ausgabe erfolgt als 32 kHz Stereo WAV.

Provenance

Detected
Sep 27, 2026
First source
huggingface
Description
Summary of the model card
Editorially reviewed
β€”