Jedrzej Kosinski

I'm having the same issue, my 4090 is generating slower/around same speed as my 3090 used to on the same machine. I am using Windows 10, with --xformers as my...

Gotcha, guess my 4090 performance will be meh until pytorch (and xformers) gets lovelace support. Those numbers were for my 4090, my 3090 was not plugged in at the time,...

If you update Advanced-ControlNet, I've just pushed a fix for the xformers thing for SparseCtrl.

ModelPatcherAndInjector is an AnimateDiff-Evolved thing, not native to ComfyUI. Based on your error logs, it looks like a change is needed to work with patch_model_lowvram. I didn't see this until...

The fix was luckily very simple and is now merged into AnimateDiff-Evolved. Update AnimateDiff-Evolved, and the issue should be fixed. Let me know if it works now! https://github.com/Kosinkadink/ComfyUI-AnimateDiff-Evolved/pull/319

Looks like ComfyUI had a new update 12 hours ago that I will need to make changes to account for, which is why you got the unpatch_weights missing thing. To...

A few seconds ago I pushed an update that should fix it - see if it works now. The change in ComfyUI code meant that any time a model containing...

I don't think the error you're getting there is relevant to the changes above - the changes don't affect tensor sizes.

This error is entirely unrelated to the one mentioned here. You can open up a new issue on AnimateDiff-Evolved, where you can also post a screenshot of your workflow.

@semaster the error you're getting is actually a bug in ComfyUI with one of the recent updates. LCM + latent batch of greater than 1 no longer works (AKA, AnimateDiff...