Skip to content

[Discussion & Proposal] Running WebUI with ROCm on Windows (tested on RX 9060 XT) #100

@cyatarow

Description

@cyatarow

The following instruction is the same content as what was posted to the "Audio Separation" server on Discord.
If there are any errors, better methods or commands, please let me know.
I would like to reflect this way in this repo if possible.


Prerequisites

Steps

  1. Download and extract the ZIP file of this repo (or if you installed git, you can also run git clone command)
  2. Open the command prompt (cmd.exe) in the extracted folder.
python -m venv venv
.\venv\Scripts\activate
pip install --no-cache-dir ^
    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
pip install --no-cache-dir ^
    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/torchaudio-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
pip install pyyaml gradio ml_collections librosa safetensors omegaconf mido beartype rotary_embedding_torch neuraloperator "PySide6-Fluent-Widgets[full]"
pip install -r requirements.txt
  1. Delete show_api=False, on line 90 of webui.py
  2. It's ready! Enter the GUI by running this:
python webui.py

My recommended batch file to enter the GUI:

@echo off
cd .\MSST-WebUI

set TORCH_ROCM_AOTRITON_ENABLE_EXPERIMENTAL=1

call .\venv\Scripts\activate
python webui.py
pause
  • "cd .\MSST-WebUI"... Please change the path to match where you installed WebUI (if it's on a different drive, you'll need "/d" option)
  • "set TORCH_ROCM_AOTRITON_ENABLE_EXPERIMENTAL=1"... increases render speed dramatically and decreases VRAM usage significantly

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions