MiniMax Music 2.0

• 🎵 Generates complete songs with synchronized vocals and instrumentation
• ✍️ Provide your own lyrics with verse/chorus song structure
• 🎯 Pairs a style/mood prompt with separate lyrical content
• 🔧 Section structure tags help control how arrangements flow
• 💬 Designed for lyric-driven, full-length sung tracks
• 🏢 Built by MiniMax as part of the minimax-music family
• ⚡ Succeeded by newer 2.5 and 2.6 releases

MiniMax

MiniMax is an AI company building generative models across multiple modalities, with a focus that spans both language understanding and audio creation. Their rapid release cadence in early 2026—delivering several new models within just a few months—reflects…

MiniMax Music 2.0 is a text-to-music model that produces full songs with AI-synthesized vocals and backing instrumentation, rather than short clips or loops. Per MiniMax's official music-generation documentation, the workflow pairs a style/mood prompt that sets genre and atmosphere with a separate lyrics input that carries the actual words, structured using familiar section markers for verses, choruses, and bridges. This separation lets creators reuse the same lyrics across different musical directions, while the structure tags guide how the composition flows from section to section.

Because the lyrics are supplied directly with verse and chorus structure, the model can generate coherent multi-section pieces that read as complete songs with singing rather than truncated fragments.

Within the catalog, MiniMax Music 2.0 sits early in the music lineage and was later followed by same-family successors MiniMax Music 2.5 and MiniMax Music 2.6, which continue MiniMax's song-generation track. It is distinct from MiniMax's text models such as MiniMax M3 and the speech-focused MiniMax Speech-02 HD.

No benchmark figures from the provider or an independent top evaluator were available, so this description is limited to documented features. For lyric-driven songwriting, soundtrack creation, and jingles, the model offers a straightforward prompt-plus-lyrics path to complete, sung tracks.