cinemasetfree.com
meta/musicgen
Generate music from a prompt or melody
Capabilities
Cost
Community model (estimated from hardware time)
Input Parameters
| Name | Type | Description | Default | Constraints |
|---|---|---|---|---|
classifier_free_guidance | integer | Increases the influence of inputs on the output. Higher values produce lower-varience outputs that adhere more closely to inputs. | 3 | — |
continuation | boolean | If `True`, generated music will continue from `input_audio`. Otherwise, generated music will mimic `input_audio`'s melody. | false | — |
continuation_end | integer | End time of the audio file to use for continuation. If -1 or None, will default to the end of the audio clip. | — | min: 0 |
continuation_start | integer | Start time of the audio file to use for continuation. | 0 | min: 0 |
duration | integer | Duration of the generated audio in seconds. | 8 | — |
input_audio | string (uri) | An audio file that will influence the generated music. If `continuation` is `True`, the generated music will be a continuation of the audio file. Otherwise, the generated music will mimic the audio file's melody. | — | — |
model_version | string | Model to use for generation | "stereo-melody-large" | stereo-melody-large stereo-large melody-large large |
multi_band_diffusion | boolean | If `True`, the EnCodec tokens will be decoded with MultiBand Diffusion. Only works with non-stereo models. | false | — |
normalization_strategy | string | Strategy for normalizing audio. | "loudness" | loudness clip peak rms |
output_format | string | Output format for generated audio. | "wav" | wav mp3 |
prompt | string | A description of the music you want to generate. | — | — |
seed | integer | Seed for random number generator. If None or -1, a random seed will be used. | — | — |
temperature | number | Controls the 'conservativeness' of the sampling process. Higher temperature means more diversity. | 1 | — |
top_k | integer | Reduces sampling to the k most likely tokens. | 250 | — |
top_p | number | Reduces sampling to tokens with cumulative probability of p. When set to `0` (default), top_k sampling is used. | 0 | — |
classifier_free_guidance integer Increases the influence of inputs on the output. Higher values produce lower-varience outputs that adhere more closely to inputs.
3 continuation boolean If `True`, generated music will continue from `input_audio`. Otherwise, generated music will mimic `input_audio`'s melody.
false continuation_end integer End time of the audio file to use for continuation. If -1 or None, will default to the end of the audio clip.
continuation_start integer Start time of the audio file to use for continuation.
0 min: 0 duration integer Duration of the generated audio in seconds.
8 input_audio string An audio file that will influence the generated music. If `continuation` is `True`, the generated music will be a continuation of the audio file. Otherwise, the generated music will mimic the audio file's melody.
model_version string Model to use for generation
"stereo-melody-large" multi_band_diffusion boolean If `True`, the EnCodec tokens will be decoded with MultiBand Diffusion. Only works with non-stereo models.
false normalization_strategy string Strategy for normalizing audio.
"loudness" output_format string Output format for generated audio.
"wav" prompt string A description of the music you want to generate.
seed integer Seed for random number generator. If None or -1, a random seed will be used.
temperature number Controls the 'conservativeness' of the sampling process. Higher temperature means more diversity.
1 top_k integer Reduces sampling to the k most likely tokens.
250 top_p number Reduces sampling to tokens with cumulative probability of p. When set to `0` (default), top_k sampling is used.
0 671ac645ce5e Updated: 2/26/2026 3.3M runs