Wan2gp suppts RDNA2 - RDNA4 GPUs. You have RDNA 1. 8GB on an AMD GPU is most likely not enough to run the models. However, maybe small models will work though.
To make it work:
- Click the menu icon of wan2gp on the main page of pinokio
- On the right side menu click Open dev mode
- At the bottom of the following page click CMD - Python Shell
- In the terminal that appears copy / paste this command:
uv pip install --pre torch torchvision torchaudio --index-url https://rocm.nightlies.amd.com/v2-staging/gfx101X-dgpu/
This is one long command. Just copy it and paste it in the terminal.
- After torch was installed, stop the terminal and try to start wan2gp-amd
Please let me know if it works. Also if you encounter any troubles using the app, post the full error you get and I can tell you if it is a limitation of your GPU or just a wrong setting. Don't try the wan2.1 model. It needs forever on AMD. Try the distilled LTX 2.3 model. Before you try to generate, go in wan2gp-amd to Configuration - Performance and set all 8 bit settings to 16 bit. Save the Preferences at the bottom. After that you can try the LTX 2.3 distilled model.
Wan2gp suppts RDNA2 - RDNA4 GPUs. You have RDNA 1. 8GB on an AMD GPU is most likely not enough to run the models. However, maybe small models will work though.
To make it work:
uv pip install --pre torch torchvision torchaudio --index-urlhttps://rocm.nightlies.amd.com/v2-staging/gfx101X-dgpu/This is one long command. Just copy it and paste it in the terminal.
Please let me know if it works. Also if you encounter any troubles using the app, post the full error you get and I can tell you if it is a limitation of your GPU or just a wrong setting. Don't try the wan2.1 model. It needs forever on AMD. Try the distilled LTX 2.3 model. Before you try to generate, go in wan2gp-amd to Configuration - Performance and set all 8 bit settings to 16 bit. Save the Preferences at the bottom. After that you can try the LTX 2.3 distilled model.