Unable to load pytorch_model.bin for CompVis--stable-diffusion-v1-4

Hi all I was trying to load CompVisā€“stable-diffusion-v1-4 (CompVis/stable-diffusion-v1-4 at main) in my jupyter notebook, but have run into some errors:

from diffusers import StableDiffusionPipeline
pipe = StableDiffusionPipeline.from_pretrained("CompVis/stable-diffusion-v1-4")
OSError: Error no file named pytorch_model.bin found in directory /home/ubuntu/.cache/hugging
face/diffusers/models--CompVis--stable-diffusion-v1-4/snapshots/249dd2d739844dea6a0bc7fc27b3c
1d014720b28/safety_checker but there is a file for Flax weights. Use `from_flax=True` to load
this model from those weights.

The import seems to run fine on a smaller compute instance I have (both have diffusers 0.12.1).

pip list | grep diffuse
diffusers                    0.12.1

On checking the folder mentioned in the error message, I indeed do not see any .bin weights file:

ubuntu@ixyz: ~/.cache/huggingface/diffusers/models--CompVis--stable-diffusion-v1-4/snapshots/249dd2d739844dea6a0bc7fc27b3c1d014720b28/safety_checker$ ls
config.json  model.safetensors

Any help will be great,
Thanks!

For now manually downloading the model file and copying it over seems like the only fix

hi @AKM-1! I unfortunately canā€™t replicate this. Could you maybe upgrade to the latest diffusers release and see if that fixes it?