Skip to content
Discussion options

You must be logged in to vote

I solved this problem by installing a nightly PyTorch build specific to my gfx:

  1. Use a venv or conda environment with Python 3.12
  2. Run pip install --index-url https://rocm.nightlies.amd.com/v2/gfx1151/ "rocm[libraries,devel]" if you need ROCm (or another gfx if this is not yours. Use rocm-info (if on Linux, you can | grep gfx)
  3. Run pip install --index-url https://rocm.nightlies.amd.com/v2/gfx1151/ --pre torch torchaudio torchvision to get the nightly torch build supporting the 8060s's gfx
  4. Make sure everything went smoothly by running python -c "import torch; print(torch.cuda.is_available())". It might say CUDA here, but if True it means that the ROCm-compiled torch build was correctly insta…

Replies: 1 comment

Comment options

You must be logged in to vote
0 replies
Answer selected by rsyue
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
1 participant