MisterChief95

Yeah you're fine hardware-wise. I didn't notice it before, but It's loading the UNET then the CLIP weights for each Lora. I tried reproducing this, but my setup only loads...

Looks like its specific to the Loras you're using. I grabbed one of the loras from your screenshot to check and saw the same thing: ``` [LORA] Loaded D:\AI\stable-diffusion-webui-forge\models\Lora\pony\Western_art_style_-_Melkor_Mancin__Riz_draws_-_combined_style-000037.safetensors for...

You may need to lower your gpu weights slider to about half at the top of the UI. Also have you tried the `Diffusion in Low Bits` options with `(fp16...

Need more info. Can you share the terminal output and model you're using?

Does it work if you try `python -m pip install triton-windows`? Triton 3.3 is for Pytorch >= 2.7.

~~Could it be because you're using the CUDA 12.6 version of Pytorch 2.7.0?~~ I think i see the issue, your pytorch and xformers CUDA versions don't match: `pytorch.version: 2.7.0+cu128` `build.torch_version:...

Some tools you can try: - [OneTrainer](https://github.com/Nerogar/OneTrainer) - [Kohya SS](https://github.com/[bmaltais/kohya_ss](https://github.com/bmaltais/kohya_ss)) You'll want to look for relevant docs/questions in those repos to learn how.

Can you share some terminal outputs? Just start forge, then run 2 generations (or as many as needed until it crashes), then copy & share the output here. Also check...

I don't think its happening to you, but one thing you can try is disable Nvidia's system memory fallback: https://nvidia.custhelp.com/app/answers/detail/a_id/5490/~/system-memory-fallback-for-stable-diffusion

Apparently lllysaviel made a post about this a while back: https://github.com/lllyasviel/stable-diffusion-webui-forge/discussions/1474 TLDR: Its caused by not enough swap (at least 40GB required supposedly). I occasionally get the `Press any key...