About this model
Promptable full-song generation with vocals, lyrics, BPM and key control. MiniMax Music 2.6 is MiniMax's latest music generation model for full vocal songs and instrumentals from text prompts.
Pricing
| Rate | RODI | USD (ref.) | Unit |
|---|---|---|---|
| Per unit | |||
| Per generation | 113.7 | ~ 0.15 | per generation (RODI/gen, USD/gen) |
RODI prices include RodiumAi markup and upstream fees. USD figures are wholesale reference rates.
Providers & pricing
| Provider | Input | Output | Cached | Free tier | Status |
|---|---|---|---|---|---|
 Runware Default All rates (1)
| - | - | - | - | No data |
Capabilities
Streaming
Tool calling
Vision
JSON mode
Reasoning
API usage
Examples
…Example response
…With lyrics
…Use cases
- Background music for videos, podcasts, ads and apps.
- Jingles and sound logos for a brand.
- Demo tracks and sketches for musicians and producers.
- Sound effects and ambiences for games and prototypes.
- Quick music options for creators who have no licence budget.
Notes
prompt(style, mood, instruments) can be up to 2000 characters.lyricsup to 3500 characters, with section tags such as[Intro],[Verse],[Chorus],[Bridge],[Outro].- Set
"instrumental": truefor a track without vocals. - To let the model write or polish lyrics from your prompt, add
"runware": {"settings": {"lyricsOptimizer": true}}. output_formatacceptsmp3(default),wav,flacandogg.seedmakes a result reproducible (0 to 1000000).- Do not send
durationunless the pricing table says the model is billed per minute: length is decided by the lyrics and the prompt.