GitLab

* Change calculation of memory total to be more accurate, allocated is actually smaller than reserved.

* Update README.md install documentation for Intel GPUs.

The previous fix didn't cover the case where the model was loaded in
lowvram mode right before.

Change torch.xpu to ipex.optimize, xpu device initialization and remove workaround for text node issue from older IPEX. (#3388)

Lower peak memory usage when changing model.

Fix case where model weights would be unloaded and reloaded.

Logging level now defaults to info. --verbose sets it to debug.

Let me know if this breaks anything.

Use fp16 text encoder weights for CPU inference to lower memory usage.

--fp8_e4m3fn-unet and --fp8_e5m2-unet are the two different formats
supported by pytorch.

Pytorch supports two variants of fp8:
--fp8_e4m3fn-text-enc (the one that seems to give better results)
--fp8_e5m2-text-enc