← Back to all generators

lucataco/ace-step

A Step Towards Music Generation Foundation Model text2music

Capabilities

Seed

Cost

Community model (estimated from hardware time)

Input Parameters

tags required string

Text prompts to guide music generation, e.g., 'epic,cinematic'

duration number

Duration of the generated audio in seconds. -1 means a random duration between 30 and 240 seconds.

Default: 60 min: 1, max: 240
granularity_scale number

Omega scale for APG guidance, or similar for other CFG types.

Default: 10 min: -100, max: 100
guidance_interval number

Guidance interval.

Default: 0.5 min: 0, max: 1
guidance_interval_decay number

Guidance interval decay.

Default: 0 min: 0, max: 1
guidance_scale number

Overall guidance scale.

Default: 15 min: 0, max: 30
guidance_type string

Guidance type for CFG.

Default: "apg"
apg cfg cfg_star
lyric_guidance_scale number

Guidance scale for lyrics.

Default: 0 min: 0, max: 10
lyrics string

Lyrics for the music. Use [verse], [chorus], and [bridge] to separate different parts of the lyrics. Use [instrumental] or [inst] to generate instrumental music

min_guidance_scale number

Minimum guidance scale.

Default: 3 min: 0, max: 100
number_of_steps integer

Number of inference steps.

Default: 60 min: 10, max: 200
scheduler string

Scheduler type.

Default: "euler"
euler heun
seed integer

Random seed. Set to -1 to randomize.

Default: -1
tag_guidance_scale number

Guidance scale for tags (text prompt).

Default: 0 min: 0, max: 10
Version: 280fc4f9ee50 Updated: 2/26/2026 107.5K runs