FastSDCPU Crashes when trying to load SDXL model

[mal9155-hash]

Whenever I try to load a SDXL model through lcm lora mode with the lcm model, it just instantly crashes.

No error log, just pipeline initialization then it just crashes and closes itself.

[monstruosoft]

Can you try running it from a console window and post the console output?.
Also, on Windows, you can try replacing the backslashes in the model path with forward slashes. Not sure if that'll work but it's worth giving it a try.

[mal9155-hash]

I restarted my computer. And unclicking "use offline cached models". It loads the model but now gets a " not enough memory" message. It might be a ram issue. I couldn't remember the exact message but it said something about CPU allocation or something similar. How much RAM is needed to run sdxl models in this program? It doesn't really say.

[monstruosoft]

SDXL requires something between 14 and 16 GB of RAM on my machine, not counting memory already used by the OS. If you're short on RAM the OS might use virtual memory on disk but, in my experience, that makes generation extremely slow and it's not really worth it.