forked from AUTOMATIC1111/stable-diffusion-webui
-
Notifications
You must be signed in to change notification settings - Fork 236
Open
Description
Checklist
- The issue exists after disabling all extensions
- The issue exists on a clean installation of webui
- The issue is caused by an extension, but I believe it is caused by a bug in the webui
- The issue exists in the current version of the webui
- The issue has not been reported before recently
- The issue has been reported before but has not been fixed yet
What happened?
Issue:
Traceback (most recent call last):
File "G:\stable-diffusion-webui-directml\launch.py", line 48, in <module>
main()
File "G:\stable-diffusion-webui-directml\launch.py", line 39, in main
prepare_environment()
File "G:\stable-diffusion-webui-directml\modules\launch_utils.py", line 618, in prepare_environment
raise RuntimeError(
RuntimeError: Torch is not able to use GPU; add --skip-torch-cuda-test to COMMANDLINE_ARGS variable to disable this check
After adding --skip-torch-cuda-test, WebUI works on the CPU. It worked before on RX 580 and RX Vega 56 (in 2023-2024), but the error appeared relatively recently.
I tried reinstalling Torch, but it didn't help.
Steps to reproduce the problem
- AMD GPU(maybe from RDNA 2)
What should have happened?
Previously, Torch could cope even without the GPU having CUDA cores.
What browsers do you use to access the UI ?
No response
Sysinfo
Console logs
venv "G:\stable-diffusion-webui-directml\venv\Scripts\Python.exe"
Python 3.10.6 (tags/v3.10.6:9c7b4bd, Aug 1 2022, 21:53:49) [MSC v.1932 64 bit (AMD64)]
Version: v1.10.1-amd-43-g1ad6edf1
Commit hash: 1ad6edf170c2c4307e0d2400f760a149e621dc38
Traceback (most recent call last):
File "G:\stable-diffusion-webui-directml\launch.py", line 48, in <module>
main()
File "G:\stable-diffusion-webui-directml\launch.py", line 39, in main
prepare_environment()
File "G:\stable-diffusion-webui-directml\modules\launch_utils.py", line 618, in prepare_environment
raise RuntimeError(
RuntimeError: Torch is not able to use GPU; add --skip-torch-cuda-test to COMMANDLINE_ARGS variable to disable this checkAdditional information
No response
Metadata
Metadata
Assignees
Labels
No labels