This document provides a comprehensive reference for the parameters available across various audio generation models in the Scenario API. Each model has a unique modelId and a set of specific parameters that can be used to control the music generation process. Understanding these parameters is crucial for effectively utilizing the API to achieve desired audio outputs.

Below, you will find detailed information for each audio model, including its modelId, the types of parameters it accepts, allowed values, default settings, and a clear description of each parameter's function.

Beatoven Music Generation

Model ID: model_beatoven-music-generation

Input	Label	Type	Default	Min	Max	Allowed Values	Notes
`prompt`	Prompt	`string`	–	–	–	–	Required. Describe the music you want to generate
`negativePrompt`	Negative Prompt	`string`	""	–	–	–	Describe instruments, moods, or styles to avoid
`duration`	Duration	`number`	90	5	150	–	Length in seconds
`refinement`	Refinement	`number`	100	10	200	–	Higher values can improve quality
`creativity`	Creativity	`number`	16	1	20	–	Higher values allow more creative interpretation
`seed`	Seed	`number`	–	0	2147483647	–	Optional reproducibility seed

Google Lyria 2

Model ID: model_lyria-2

Input	Label	Type	Default	Notes
`prompt`	Prompt	string	–	Required. Up to 2048 chars
`negativePrompt`	Negative Prompt	string	–	Excludes elements from generation
`seed`	Seed	number	–	Optional reproducible seed

Meta MusicGen

Model ID: model_meta-musicgen

Input	Label	Type	Default	Min	Max	Allowed Values	Notes
`modelVersion`	Model Version	select	stereo-melody-large	–	–	stereo-melody-large, stereo-large, melody-large, large
`prompt`	Prompt	string	–	–	–	–	Required if no input_audio
`inputAudio`	Input Audio	file	–	–	–	–	optional conditioning
`duration`	Duration	number	8	1	30	–	seconds
`continuation`	Continuation	boolean	false	–	–	–	continues from input_audio
`continuationStart`	Start	number	0	0	–	–	start time (s)
`continuationEnd`	End	number	–	0	–	–	defaults to end
`multiBandDiffusion`	Multi Band Diffusion	boolean	false	–	–	–	only for non-stereo models
`normalizationStrategy`	Normalization Strategy	select	loudness	–	–	loudness, clip, peak, rms
`temperature`	Temperature	number	1	–	–	–	controls diversity
`classifierFreeGuidance`	Guidance	number	3	0	10	–	higher = more faithful
`seed`	Seed	number	–	–	–	–	optional RNG seed

Minimax

Minimax Music 2.0

Model ID: model_minimax-music-2-0

Input	Label	Type	Default	Min	Max	Allowed Values	Notes
`prompt`	Prompt	`string`	None	10	300	None	Required. Description of style, mood, scenario
`lyrics`	Lyrics	`string`	None	10	3000	None	Required. Use \n for line breaks. Structure tags like [Intro], [Verse], [Chorus], [Outro] are allowed
`audio_setting.sample_rate`	Sample rate	`number`	44100	None	None	8000, 16000, 22050, 24000, 32000, 44100	Controls audio sample rate
`audio_setting.bitrate`	Bitrate	`number`	256000	None	None	32000, 64000, 128000, 256000	Output audio bitrate

Minimax Music 1.5

Model ID: model_minimax-music-1-5

Input	Label	Type	Default	Allowed Values	Notes
`prompt`	Prompt	string	–	–	10-300 characters required
`lyrics`	Lyrics	string	–	–	10-600 characters
`sampleRate`	Sample Rate	number	44100	16000, 24000, 32000, 44100
`bitrate`	Bitrate	number	256000	32000, 64000, 128000, 256000

Minimax Music 01

Model ID: model_minimax-music-01

Input	Label	Type	Default	Allowed Values	Notes
`lyrics`	Lyrics	string	""	–	required. Supports newline and ## for accompaniment
`songFile`	Song File	file	–	–	must be >15s
`voiceFile`	Voice File	file	–	–	required if lyrics given
`instrumentalFile`	Instrumental File	file	–	–	instrumental reference
`sampleRate`	Sample Rate	number	44100	16000, 24000, 32000, 44100
`bitrate`	Bitrate	number	256000	32000, 64000, 128000, 256000

MMAudio

MM Audio 2

Model ID: model_mm-audio-2

Input	Label	Type	Default	Min	Max	Allowed Values	Notes
`prompt`	Prompt	`string`	None	None	2048	None	Required. Text prompt describing the desired audio
`video`	Video	`assetId`	None	None	None	None	Required. Video used to synchronize audio generation
`negativePrompt`	Negative Prompt	`string`	None	None	2048	None	Optional. Sounds to avoid
`duration`	Duration	`number`	8	1	30	None	Output duration in seconds
`numSteps`	Steps	`number`	25	4	50	None	Number of generation steps
`cfgStrength`	Guidance	`number`	4.5	1	20	None	Higher values follow the prompt more closely
`maskAwayClip`	Mask Away Clip	`boolean`	false	None	None	None	Masks away certain sounds from the clip
`seed`	Seed	`number`	None	0	65535	None	Optional reproducibility seed

MM Audio 2 Text to Audio

Model ID: model_mm-audio-2-t2a

Input	Label	Type	Default	Min	Max	Allowed Values	Notes
`prompt`	Prompt	`string`	None	None	2048	None	Required. Text description of the audio
`negativePrompt`	Negative Prompt	`string`	None	None	2048	None	Optional. Sounds to avoid
`duration`	Duration	`number`	8	1	30	None	Output duration in seconds
`numSteps`	Steps	`number`	25	4	50	None	Number of generation steps
`cfgStrength`	Guidance	`number`	4.5	1	20	None	Higher values follow the prompt more closely
`maskAwayClip`	Mask Away Clip	`boolean`	false	None	None	None	Masks away certain sounds
`seed`	Seed	`number`	None	0	65535	None	Optional reproducibility seed

MMAudio 1

Model ID: model_mm-audio

Input	Label	Type	Default	Min	Max	Notes
`prompt`	Prompt	string	–	–	–	required
`video`	Video	file	–	–	–	required
`negativePrompt`	Negative Prompt	string	–	–	–	–
`duration`	Duration	number	8	1	30	seconds
`numSteps`	Steps	number	25	4	50	–
`cfgStrength`	Guidance	number	4.5	1	20	higher = closer to prompt
`seed`	Seed	number	–	-1	–	-1 or blank = random