Nodes/ComfyUI-INT8-Fast-Fork/Apply LoRA Stack (Quantized)
ComfyUI Node

Apply LoRA Stack (Quantized)

Patch any number of configured LoRAs into a quantized or floating-point diffusion MODEL.

By SparknightLLC·Created 5 months ago·Updated 5 days ago· 56
Apply LoRA Stack (Quantized)
  • model
  • loras
  • MODEL
modeStochastic
Categoryloaders

Inputs (3)

NameTypeDefaultDescription
modeCOMBOStochasticStandard uses ComfyUI patching. Stochastic requantizes patched weights. Dynamic preserves runtime deltas on INT8/W4A4 and warns before using Standard fallback on W4A8.
modelMODELQuantized or floating-point diffusion model to receive the LoRAs.
lorasoptCOMFY_AUTOGROW_V3Connect LoRA Stack Entry (Quantized) outputs; another input appears as each one is connected.

Outputs (1)

NameTypeDescription
MODELMODEL