[BUG REPORT] RuntimeError: CUTLASS submodule not found.

[lj-thirdeye]

I follow the steps to install the webui on my Windows system. When I run "run.bat", I met this error:

STDERR: error: subprocess-exited-with-error

python setup.py egg_info did not run successfully. exit code: 1

[9 lines of output] No CUDA runtime is found, using CUDA_HOME='C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.8' Traceback (most recent call last): File "", line 2, in File "", line 34, in File "C:\Users\jlmAurora\AppData\Local\Temp\pip-install-vfd6n54r\xformers_d33f228d057c4d9f8f92c2cbeb4093cf\setup.py", line 239, in ext_modules=get_extensions(), File "C:\Users\jlmAurora\AppData\Local\Temp\pip-install-vfd6n54r\xformers_d33f228d057c4d9f8f92c2cbeb4093cf\setup.py", line 157, in get_extensions raise RuntimeError( RuntimeError: CUTLASS submodule not found. Did you forget to run git submodule update --init --recursive ? [end of output]

note: This error originates from a subprocess, and is likely not a problem with pip. error: metadata-generation-failed

Encountered error while generating package metadata.

See above for output.

note: This is an issue with the package mentioned above, not pip. hint: See above for details.

After I run pip install catlass and git submodule update --init --recursive, it still returned the same error. Then I also deleted and reinstalled everything, but the error is still there. What's going on?

[gitmylo]

It might have something to do with recent changes to pytorch, as the new default CUDA version for pytorch is now 12.1 instead of 11.8. I'll lock it back to the older version, as that might fix this issue as well.

[lj-thirdeye]

It might have something to do with recent changes to pytorch, as the new default CUDA version for pytorch is now 12.1 instead of 11.8. I'll lock it back to the older version, as that might fix this issue as well.

Okay I see. New versions do have the ability to cause issues. Thanks for the reply!