Morph MORPH
SIGN IN SIGN UP
dhruvhead / ComfyUI-QuantFunc UNCLAIMED
0 0 2 Python

COMMITS

main
April 4, 2026
Q
feat: add Base Model Auto Loader with Download node
QuantFunc committed
a28aa87
Q
chore: remove auto-creation of lora directory
QuantFunc committed
3e3ce54
Q
fix: run git pull before module imports so new code takes effect immediately
QuantFunc committed
b1dfa00
Q
refactor: use standard ComfyUI model directories for auto loaders
QuantFunc committed
77c77e2
Y
fix: add fallback for base model repo list and filter "None" api_key
yejianhui committed
859f1b2
Q
feat: add Base Series Model Auto Loader node for QuantFunc series base models
QuantFunc committed
df40659
Q
feat: auto-create lora and transformer directories on startup
QuantFunc committed
b06dc41
Q
feat: add Transformer Auto Loader node to scan models/QuantFunc/transformer/
QuantFunc committed
7c45a48
Q
feat: support recursive subfolder scanning in LoRA Auto Loader
QuantFunc committed
f2a5319
Q
fix: use org/repo directory structure for base model downloads
QuantFunc committed
f02b67f
Q
feat: add LoRA Auto Loader node to scan and load LoRA from models/QuantFunc/lora/
QuantFunc committed
57613a3
Q
feat: add Base Model Auto Loader node and auto git pull on startup
QuantFunc committed
e8e5377
Q
fix: handle non-dict worker response to prevent AttributeError on DLL load failure
QuantFunc committed
c4ea01e
April 3, 2026
Q
feat: add Prequant and PrecisionConfig auto loader nodes, remove scheduler_config from AutoLoader
QuantFunc committed
e019452
Q
fix: detect and re-download corrupt vision_encoder weights for Qwen-Edit
QuantFunc committed
50b619e
Q
feat: auto-download missing .so dependencies from ModelScope on Linux
QuantFunc committed
951d00a
Q
fix: distinguish worker crash from timeout, show actual error message
QuantFunc committed
6eec8c6
Q
fix: increase pipeline create timeout to 1800s for lighting warmup
QuantFunc committed
30cca16
Q
fix: restore precision_config, prequant_weights, fused_mod to ModelLoader
QuantFunc committed
69f192b
Q
fix: always pass sampler in options_json to prevent stale state
QuantFunc committed
545d341
Q
feat: add sampler and eta options to Generate node
QuantFunc committed
bb4426e
March 31, 2026
Q
refactor: remove unused prequant_weights, precision_config, fused_mod options
QuantFunc committed
5dea735
Q
fix: only unload pipeline when VRAM is actually insufficient
QuantFunc committed
8d2f7b8
March 30, 2026
Q
docs: add CUDA 12 installation instructions for Linux
QuantFunc committed
94431e9
Q
fix: forward **kwargs in free_memory hook for ComfyUI compatibility
QuantFunc committed
09de34c
Q
docs: add Easy Gen beginner tutorial with auto-download workflow
QuantFunc committed
f2fcd74
Q
feat: ModelAutoLoader node with auto-download from HuggingFace/ModelScope
QuantFunc committed
926a405
March 29, 2026
Q
fix: auto-download dependency DLLs when worker fails to load on Windows
QuantFunc committed
5802e4c
Q
feat: auto-install modelscope SDK + direct HTTP download fallback
QuantFunc committed
efc676b
Q
fix: fast-fail when SM120+ GPU detected with CUDA 12
QuantFunc committed
b56972f