Creates a native ACE-Step 1.5 Inpaint/Repaint task with trained source context, per-step preservation, boundary blending, and waveform splicing. — ComfyUI node…

An all-in-one node for ComfyUI that implements AceStep 1.5 SFT (Supervised Fine-Tuning), a high-quality music generation model. This node replicates the full functionality of the official Gradio pipeline, offering fine control over audio synthesis parameters.

ComfyUI-AceStep_SFT

source

start_seconds

end_seconds

repaint_mode

repaint_strength

reference_audio

Creates a native ACE-Step 1.5 Inpaint/Repaint task with trained source context, per-step preservation, boundary blending, and waveform splicing.

Name	Type	Default	Description
source	AUDIO,LATENT		Audio/latent to preserve outside the selected time range.
start_seconds	FLOAT	0.000–600	Start of the region to regenerate.
end_seconds	FLOAT	-1.00-1–600	End of the region to regenerate. -1 uses the end of the source.
repaint_mode	COMBO	balanced	Conservative preserves source context during more diffusion steps; aggressive allows more change.
repaint_strength	FLOAT	0.500–1	Balanced mode only: 0 = maximum preservation, 1 = maximum freedom.
reference_audioopt	AUDIO,LATENT		Optional independent timbre/style reference for the regenerated region.

AceStep 1.5 SFT Inpaint / Repaint

Inputs (6)

Outputs (1)