manotroll
commented
2 months ago For me, this tip did the trick https://github.com/vladmandic/automatic/issues/3318#issuecomment-2211766345
and do not use spaces in the user name this helps or use python in the all users version
Issue Description
I gave up it's beyond my knowledge hope you guys could help me to solve this problem it's seem rocm is the culcript why sd next render as cpu unable to load the models due to rocm version act as a none i'm using rx6700xt sometimes try to reinstall driver, hip sdk and pasted optimized lib gfx31 and delete venv folder the problems still won't perish.
Version Platform Description
00:48:37-493718 INFO Starting SD.Next 00:48:37-495717 INFO Logger: file="C:\Users\R U S T Y\AMDSDXL\sdnext.log" level=DEBUG size=65 mode=create 00:48:37-497718 INFO Python version=3.11.8 platform=Windows bin="C:\Users\R U S T Y\AMDSDXL\venv\Scripts\python.exe" venv="C:\Users\R U S T Y\AMDSDXL\venv" 00:48:37-651718 INFO Version: app=sd.next updated=2024-06-24 hash=94f6f0db branch=master url=https://github.com/vladmandic/automatic/tree/master ui=main 00:48:38-493716 INFO Latest published version: 081c19fc122c6c8e60fcddfc37917e8107f65290 2024-07-01T08:20:42Z 00:48:38-507716 INFO Platform: arch=AMD64 cpu=AMD64 Family 25 Model 33 Stepping 2, AuthenticAMD system=Windows release=Windows-10-10.0.22631-SP0 python=3.11.8 00:48:38-511716 DEBUG Setting environment tuning 00:48:38-512716 INFO HF cache folder: C:\Users\R U S T Y.cache\huggingface\hub 00:48:38-514716 DEBUG Torch allocator: "garbage_collection_threshold:0.80,max_split_size_mb:512" 00:48:38-515716 DEBUG Torch overrides: cuda=False rocm=False ipex=False diml=False openvino=False 00:48:38-516716 DEBUG Torch allowed: cuda=True rocm=True ipex=True diml=True openvino=True 00:48:38-524716 INFO Python version=3.11.8 platform=Windows bin="C:\Users\R U S T Y\AMDSDXL\venv\Scripts\python.exe" venv="C:\Users\R U S T Y\AMDSDXL\venv" 00:48:38-531716 INFO AMD ROCm toolkit detected 00:48:38-635719 DEBUG ROCm agents detected: [] 00:48:38-654718 DEBUG ROCm version detected: None 00:48:38-655717 WARNING ZLUDA support: experimental 00:48:38-687718 INFO Using ZLUDA in C:\Users\R U S T Y\AMDSDXL.zluda 00:48:38-688718 INFO Install: package="torch==2.3.0 torchvision --index-url https://download.pytorch.org/whl/cu118" 00:48:38-688718 DEBUG Running: pip="install --upgrade torch==2.3.0 torchvision --index-url https://download.pytorch.org/whl/cu118 "
Relevant log output
No response
Backend
Diffusers
Branch
Master
Model
SD-XL
Acknowledgements