⚙️ GRAG Advanced Sampler v3.0 is a ComfyUI node in the GRAG/v3.0 category from ComfyUI-GRAG-ArchAi3D. Inputs: model, positive, negative, latent_image. Outputs…

Advanced GRAG (Guided Region-Adaptive Guidance) implementation for ComfyUI. Features: Simple Controller (3 parameters) for beginners, Unified Controller (25+ parameters) for experts, Advanced Sampler with v2.2.1 contamination fix, Preset Manager with 54 curated presets, per-layer control, adaptive schedules, and multi-resolution support.

ComfyUI-GRAG-ArchAi3D

model

positive

negative

latent_image

seed

steps

sampler_name

scheduler

denoise

debug_mode

GRAG_Advanced_Sampler

samples

⚙️ GRAG Advanced Sampler v3.0

Name	Type	Default	Description
model	MODEL		The diffusion model used for denoising
positive	CONDITIONING		Positive conditioning (should contain GRAG metadata from Unified Controller)
negative	CONDITIONING		Negative conditioning
latent_image	LATENT		Input latent to denoise
seed	INT	00–18446744073709550000	Random seed for noise generation
steps	INT	201–10000	Number of denoising steps
cfg	FLOAT	8.00–100	Classifier-Free Guidance scale
sampler_name	COMBO		Sampling algorithm to use
scheduler	COMBO		Noise schedule for denoising
denoise	FLOAT	1.000–1	Denoising strength (1.0 = full denoise)
debug_mode	BOOLEAN	false	Enable debug logging (attention layer info, performance metrics)

⚙️ GRAG Advanced Sampler v3.0

Inputs (11)

Outputs (1)