Nodes/ComfyUI-PolyhedronLoRAStack/⬡ Polyhedron LoRA Engine
ComfyUI Node

⬡ Polyhedron LoRA Engine

Applies engine-class LoRAs (Lightning, Turbo, LCM, FusionX, CausVid and the like) ahead of the creative stack. These change HOW the model computes - the inference trajectory - rather than WHAT it depicts, which is why they get their own node: a flat list, one global merge mode, no groups to reason about.

By PolyhedronAI·Created 3 months ago·Updated 17 days ago· 2
⬡ Polyhedron LoRA Engine
  • model
  • clip
  • MODEL
  • CLIP
  • debug_info
engine_config{"rows":[],"mode":"SEQ"}
CategoryPolyhedron/Loaders

Inputs (3)

NameTypeDefaultDescription
modelMODEL
clipoptCLIP
engine_configoptSTRING{"rows":[],"mode":"SEQ"}

Outputs (3)

NameTypeDescription
MODELMODEL
CLIPCLIP
debug_infoSTRING