Skip to main content

MusicCardConfig

Model truth for bounded text-to-music generation, separate from speech.

familyMusicModelFamily (string)required

Architecture family used by the runner's fixed option translator.

Possible values: [minimax_music3, ace_step_1_5]

lyricsMusicLyricRequirement (string)required

Whether lyrics are required, permitted, or unsupported.

Possible values: [required, optional, unsupported]

minSecondsMinseconds (integer)required

Shortest generation target accepted for this artifact.

Possible values: > 0

maxSecondsMaxseconds (integer)required

Longest generation target accepted for this artifact, at most 120 seconds.

Possible values: > 0

languageModelGguf object

MiniMax language-model component selected by this exact artifact.

anyOf
string
rvqDepthDecoderGguf object

MiniMax RVQ decoder component selected by this exact artifact.

anyOf
string
flowTransformerGguf object

MiniMax flow-transformer component selected by this exact artifact.

anyOf
string
MusicCardConfig
{
"family": "minimax_music3",
"lyrics": "required",
"minSeconds": 0,
"maxSeconds": 0,
"languageModelGguf": "string",
"rvqDepthDecoderGguf": "string",
"flowTransformerGguf": "string"
}