lewington
lewington
I have the exact same issue. I think that this repo is just too old and now out of date (looks like 11 months since last update). I'm just going...
it looks like you can probably use the "use_auth_token" keyword argument, e.g. ```python StableDiffusionPipeline.from_pretrained(model_loc, use_auth_token="YOUR_TOKEN") ``` If this works, it probably means that huggingface is having trouble picking up the...
I spent like 60 minutes googling this yesterday, I checked a bunch of different forums and the consensus seems to be that due to the fact that SD was trained...
I'm going to start work on this right now
@patrickvonplaten after some research I think I understand the issue. The images are currently stored in a low-precision format (e.g. png) which prevents us from testing with any greater precision...
Hey, yes I am currently working on this in fact!
@patrickvonplaten I made a PR adding the files: https://huggingface.co/datasets/hf-internal-testing/diffusers-images/discussions/2
yt_dlp works for me, but it took me a little while to work out that it only works if you give it video urls formatted like `https://www.tiktok.com/@pdspopculture/video/7285771356795342123` as opposed to...
and I also
you need to move the MoE to cuda ```python # init moe on CPU moe = MoE( dim = 768, num_experts = 32, # increase the experts (# parameters) of...