Nodes/RocM Ninodes/ROCm KSampler
ComfyUI Node

ROCm KSampler

KSampler with auto-detected architecture and model optimizations

By iGavroche·Created 10 months ago·Updated about a month ago· 37
ROCm KSampler
  • model
  • positive
  • negative
  • latent_image
  • LATENT
seed0
steps20
cfg8.0
sampler_name
scheduler
denoise1.00
optimize_for_videofalse
precision_modeauto
compatibility_modefalse
CategoryROCm Ninodes/Sampling

Inputs (13)

NameTypeDefaultDescription
modelMODELThe model to sample from
seedINT00–18446744073709550000Random seed for generation
stepsINT201–10000Number of sampling steps
cfgFLOAT8.00–100Classifier-free guidance scale
sampler_nameCOMBOSampling algorithm to use
schedulerCOMBOScheduler for noise timesteps
positiveCONDITIONINGPositive conditioning
negativeCONDITIONINGNegative conditioning
latent_imageLATENTLatent image to sample from
denoiseFLOAT1.000–1Denoising strength
optimize_for_videooptBOOLEANfalseDisable previews/progress for multi-frame latents
precision_modeoptCOMBOautoROCm precision hint (no-op on CUDA/CPU)
compatibility_modeoptBOOLEANfalseForce pure stock behavior

Outputs (1)

NameTypeDescription
LATENTLATENT