MiniMax Music 3

Run it

The smallest request that works: the required fields only.

curl https://api.percify.io/v3/playground/v1/run \
  -H "Authorization: Bearer $PERCIFY_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "modelId": "minimax-music-3",
    "input": {
      "prompt": "A red fox running through fresh snow at sunrise",
      "lyrics": "A red fox running through fresh snow at sunrise"
    }
  }'

1. EstimatePOST /estimate with the same body returns the price and charges nothing.
2. RunPOST /run returns an id at once, with status processing.
3. CollectGET /generations/{id}?wait=55 returns the file URLs, or pass a webhook.

Parameters

6 fields in input, required first.

• lyricsRequired
  string

  The lyrics to sing. Structure tags such as [intro], [verse], [pre-chorus], [chorus], [post-chorus], [bridge], [instrumental], [solo] and [outro] must each be on their own line; text on the same line as a leading tag is dropped by the model's input contract.

• promptRequired
  string

  Music description: style, mood, vocals, instrumentation and arrangement. For precise control use a Structured Caption with global metadata (genre, BPM, key, emotional progression), vocal details, and a section-by-section arrangement.

• seedOptional
  string

  Random seed for reproducibility. If not provided, a random seed will be used.

• durationOptional
  numberDefault 60

  Upper bound on the generated audio length in seconds. The model may stop earlier; the actual duration is returned in the output.

• guidance_scaleOptional
  numberDefault 1.7

  Classifier-free guidance scale of the flow-matching stage.

• num_inference_stepsOptional
  integerDefault 30

  Number of flow-matching Euler steps per 8-second denoising chunk. More steps improve quality at the cost of speed.