ROCM Environment Crashes with "Access Violation" Exception

[saddam213]

The ROCm Python environment crashes with a memory access violation when attempting to run diffusion.

The model loads correctly onto the device, but as soon as inference begins, a memory error occurs that terminates the entire application.

From the .NET side, I’m unable to catch any meaningful debug information, and I have no idea how to debug/diagnose the issue from the Python side. (python n00b)

Requirements Used

[
  "https://repo.radeon.com/rocm/windows/rocm-rel-7.1.1/rocm_sdk_core-0.1.dev0-py3-none-win_amd64.whl",
  "https://repo.radeon.com/rocm/windows/rocm-rel-7.1.1/rocm_sdk_devel-0.1.dev0-py3-none-win_amd64.whl",
  "https://repo.radeon.com/rocm/windows/rocm-rel-7.1.1/rocm_sdk_libraries_custom-0.1.dev0-py3-none-win_amd64.whl",
  "https://repo.radeon.com/rocm/windows/rocm-rel-7.1.1/rocm-0.1.dev0.tar.gz",
  "https://repo.radeon.com/rocm/windows/rocm-rel-7.1.1/torch-2.9.0+rocmsdk20251116-cp312-cp312-win_amd64.whl",
  "https://repo.radeon.com/rocm/windows/rocm-rel-7.1.1/torchvision-0.24.0+rocmsdk20251116-cp312-cp312-win_amd64.whl",
  "typing==3.7.4.3",
  "wheel==0.45.1",
  "transformers==4.57.3",
  "accelerate==1.12.0",
  "diffusers==0.36.0",
  "protobuf==6.33.2",
  "sentencepiece==0.2.1",
  "ftfy==6.3.1",
  "scipy==1.16.3",
  "peft==0.18.0",
  "hf-xet==1.2.0",
  "torchsde==0.2.6"
]