Joel Simon

Manually installing from the feature/python3 branch fixed this error for me on python3.6.

Hi @ccssu , I tried that before but ran into this error. ## Code ``` import os import torch from diffusers import StableDiffusionXLPipeline from diffusers.utils import load_image as _load_image from...

@ccssu Thanks, unfortunately that returns the same error. Diffusers just build from source - 0.28.0.dev0 - I tried earlier on 0.27.0 , torch==1.13.1

In case its helpful, I add a print there... ``` def head_to_batch_dim(self, tensor, out_dim=3): head_size = self.heads print("tensor.shape", tensor.shape) batch_size, seq_len, dim = tensor.shape tensor = tensor.reshape(batch_size, -1, head_size, dim...

adding these two lines got the unet to compile and full pipeline to compile as well. Almost a fix but then in my app with multiple ipadapters it ignores all...

@ccssu ah gotcha, torch >= 2.0 did fix it, thanks! It's not working for multiple ip adapters but I don't think that's supported yet, right? Looking forward to that, thanks.

@ccssu Thanks, that example does now work! My code with some other modifications does not so I will investigate further and follow up.

I am on OSX 10.7.4 as well as experiencing an identical issue. Latest Nite, OpenNi and avin2 as well. Programs will run for 10-20 seconds before crashing with the same...

FF seems to have made some big improvements. I wish I had the time for this now but unfortunately I don't :(

Just added GPL (V2), I thought I had added that before, oops. That would be awesome if you're willing to do that, it shouldn't be very hard. I think FBG...