Disty0

Use Linux Kernel 6.10.

Your drivers (Level Zero and Intel Compute Runtime) are probably too old for IPEX. IPEX is unable use the GPU.

Closing this as the issue thread is idle for months. Feel free to open a new issue if this is still happening on latest Pytorch. You can also try the...

NNCF lists: `pyparsing

A driver crash. Use Linux if you want a stable ARC.

Closing this as the issue thread is idle for months. Feel free to open a new issue if this is still happening on latest Pytorch.

Load the Stable Cascade base (the decoder) from Huggingface as the main model first, then load a Unet (prior) as the Unet. Put the Unet safetensors in Unet folder and...

Hijacking this issue, this slowness also happens every time when you move the models between cpu and gpu on my end. Edit: Slowness after the first move on my end...

Done.

Same issue here with Kohya SS Stable Diffusion training. I am training a 512 dim SD 1.5 Lora at 1024x1536, Batch Size 1. Kohya SS: https://github.com/kohya-ss/sd-scripts/ IPEX specific hijacks to...