Jelle De Loecker

@ChrisLane I was using MultiMC to set the custom glfw library via the java arguments config, indeed :) @ninja- Once I installed libdecoration it worked fine! Well, the actual decoration...

@ninja- I'm on Manjaro. It does affect the FPS quite a bit, I'm just guessing that's due to the fractional scaling + screen capturing + BSL shaders. But I also...

Still works fine in Gnome 3.38 **and** capturing the window using `obs-xdg-portal` no longer crashes Gnome, so I guess that was a gnome/mutter specific issue. Yay!

According to [this thread on the pytorch forum](https://discuss.pytorch.org/t/missing-keys-unexpected-keys-in-state-dict-when-loading-self-trained-model/22379/3), the pretrained model was saved using something called `nn.DataParallel`, but the `eval.py` script will only do something with that if cuda is...

 My current solution broke Tweakaroo's fast placement. I think I can add a check to see which hand is doing the interaction

@RoboTricker The last dev build of ProtocolLib happily runs on 1.14.4, anything we can help you with?

`wrapperLength` is more like a global setting, not something that you can specify for code run using the `vm` module. If being able to dynamically set the wrapper length using...

I'm using the llama 3 colab file, but on a vast.ai server: ``` !pip install "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git" !pip install --no-deps "xformers

I'm getting the same error on Ubuntu, so that's not it :/

I'm getting this too. Same with Readarr.