ComfyUI Node: DiT CFG-FSG/DCW Patch
Run ComfyUI workflows without the setup
No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.
Category
model_patches
Inputs
model MODEL
enabled BOOLEAN
dcw_mode
- off
- manual
- auto
dcw_lambda FLOAT
dcw_band_mask
- LL
- all
- HH
- LH+HL+HH
dcw_calibrator
- (auto-download default)
smc_cfg BOOLEAN
adaptive_smc_alpha FLOAT
smc_cfg_lambda FLOAT
cfgpp BOOLEAN
cfgpp_lambda FLOAT
fsg BOOLEAN
fsg_band_lo FLOAT
fsg_band_hi FLOAT
fsg_k INT
fsg_d_sigma FLOAT
fsg_gamma FLOAT
replace_existing_cfg BOOLEAN
steps INT
cfg FLOAT
sampler_name
- euler
- euler_cfg_pp
- euler_ancestral
- euler_ancestral_cfg_pp
- heun
- heunpp2
- exp_heun_2_x0
- exp_heun_2_x0_sde
- dpm_2
- dpm_2_ancestral
- lms
- dpm_fast
- dpm_adaptive
- dpmpp_2s_ancestral
- dpmpp_2s_ancestral_cfg_pp
- dpmpp_sde
- dpmpp_sde_gpu
- dpmpp_2m
- dpmpp_2m_cfg_pp
- dpmpp_2m_sde
- dpmpp_2m_sde_gpu
- dpmpp_2m_sde_heun
- dpmpp_2m_sde_heun_gpu
- dpmpp_3m_sde
- dpmpp_3m_sde_gpu
- ddpm
- lcm
- ipndm
- ipndm_v
- deis
- res_multistep
- res_multistep_cfg_pp
- res_multistep_ancestral
- res_multistep_ancestral_cfg_pp
- gradient_estimation
- gradient_estimation_cfg_pp
- er_sde
- seeds_2
- seeds_3
- sa_solver
- sa_solver_pece
- ddim
- uni_pc
- uni_pc_bh2
- er_sde_cns
scheduler
- simple
- sgm_uniform
- karras
- exponential
- ddim_uniform
- beta
- normal
- linear_quadratic
- kl_optimal
denoise FLOAT
clip CLIP
positive CONDITIONING
Outputs
MODEL
Extension: SpectrumKSampler
Spectrum: training-free diffusion sampling acceleration via Chebyshev polynomial feature forecasting. Drop-in KSampler replacement that skips transformer blocks on predicted steps for ~2-3x speedup.
Authored by sorryhyun
Looking for a different node?
More nodes in SpectrumKSampler
Other model_patches nodes
- Add Magic Clothing Attention
- ApplyResAdapterUnet
- Contextual Repulsion (Diversity)
- DCW + CWM + SMC Model Patch
- Excellent attention
- Extend Clip Vision Input Size
- DIT Deep Debiaser (FLUX.2 Klein — Verified)
- VAE Deep Debiaser (Flux 2 Klein — 125 Tensors)
- FreeUExtreme
- FreeUExtremeConfig
- InjectTensorHashLog
- interrupt on NaN
Run ComfyUI workflows without the setup
No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.