All models
audio generation
MiniMax Music 2.6
MiniMax Music 2.6 is a text-to-music model available in Gobostock. From a description of genre, mood, tempo and instruments it composes a complete song with vocals or a pure instrumental. Lyrics can be written by the model, supplied by you with section tags such as Verse, Chorus and Bridge, or left out entirely, and each generation returns a full track, usually one to two and a half minutes long and longer with more lyrics, for a single flat price.
The audio tool calculates the final credit quote from the selected operation and output settings. The quote shown in the tool is the one you pay.
No sample published yet.
Best for
- Original background music for videos, podcasts and ads
- Jingles and short theme songs with custom lyrics
- Quick song drafts to test a melody idea before production
Known limits
- Song length is chosen by the model and cannot be set exactly.
- A song takes one to three minutes to generate.
- It does not imitate real artists or cover existing songs; describe the sound instead.
Capabilities
- Full songs with vocals or pure instrumentals from a text description
- Your own lyrics with section tags, or lyrics written by the model
- Optional lyric polishing for meter and rhyme
- MP3 or WAV output
Inputs
- Song description (genre, mood, tempo, instruments, voice)
- Optional lyrics with section tags
- Vocals or instrumental choice
- Optional seed
Outputs
- Generated song (MP3 or WAV)
Compare