Dhruv Nair

@yiyixuxu Anything specific you wanted to address in this current PR?

Oh wait. Nvm I haven't addressed those initial comments. Will wrap this up.

Hi @ballenvironment we don't have pipeline level single file yet for Stable Cascade. Can you point us to a single file checkpoint that exists for the entire SD Cascade pipeline?...

@ballenvironment Following up here to do you have a checkpoint we can test with for loading the Prior as single file?

@Beinsezii currently `from_single_file` is also used to load models in their original format. I realise the name can be a bit confusing in that regard. We're working to move all...

@dengfenglai321 I tried running the following snippet with version 0.26.3 and the error is still raised. ```python from diffusers import ControlNetModel, StableDiffusionXLControlNetInpaintPipeline controlnet = ControlNetModel.from_pretrained("diffusers/controlnet-canny-sdxl-1.0") pipe = StableDiffusionXLControlNetInpaintPipeline.from_single_file( "https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/blob/main/sd_xl_base_1.0.safetensors", controlnet=controlnet,...

If the argument is added to the pipeline and is only a pipeline argument then that wouldn't be a discrepancy. What we want is to avoid configuring models via pipeline...

Hi @Teriks The patch size for the decoder is configured to 2 and the embedding layer is configured based on the patch size, which is what seems to be causing...

@crystal-kur The traceback seems to indicate an issue with the torch install. Can you try installing torch 2.2. in a fresh virtualenv and run again? Once you have created this...

@saeedkhanehgir Can you share a code example that produces this error? As well as the full traceback. Currently the maximum supported length for a prompt is SD Cascade is 77...