CoreyCorza
3c67b84f31
Krea 2 (and other gated DiTs) put multiplicative sigmoid gate projections (attn.gate / mlp.gate) in every block - ~25% of a Krea 2 LoRA. Their edits have outsized, compounding effect (stacked LoRAs fighting -> deformities) but are modest in norm, so tame_layers misses them. gate_strength targets them by name and scales just those layers. Wired into both nodes; non-gate layers left untouched. Bump 1.2.0.
19 lines
702 B
TOML
19 lines
702 B
TOML
[project]
|
|
name = "comfyui-lora-loader"
|
|
description = "A drop-in ComfyUI LoRA loader with optional spectral cleanup (energy trimming + hot-layer taming) to reduce artifacts from stacked LoRAs and turbo/distilled models."
|
|
version = "1.2.0"
|
|
license = { text = "MIT" }
|
|
readme = "README.md"
|
|
requires-python = ">=3.9"
|
|
dependencies = []
|
|
|
|
[project.urls]
|
|
Repository = "https://github.com/CoreyCorza/comfyui-lora-loader"
|
|
|
|
# ComfyUI Registry metadata (https://docs.comfy.org/registry/overview).
|
|
# Set PublisherId to your registry publisher handle before `comfy node publish`.
|
|
[tool.comfy]
|
|
PublisherId = ""
|
|
DisplayName = "Corza LoRA Loader (Clean)"
|
|
Repository = "https://github.com/CoreyCorza/comfyui-lora-loader"
|