stabilityai/stable-diffusion-2-1 is missing diffusion_pytorch_model.fp16.safetensors

[Gr6800]

[a-r-r-o-w]

If possible on your end, you can create a local copy of the fp16 safetensors:

import torch
from diffusers import DiffusionPipeline

model_id = "stabilityai/stable-diffusion-2-1"
pipe = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16)
pipe.save_pretrained(model_id, variant="fp16")

This will create a folder with vae, unet, etc. with the files saved in safetensors format. To convert to a single safetensors file in the original SD format, you can run this for SDXL and this for SD1.5. I'm not sure if the latter would work with SD2.1 as well because of different model architecture. If it doesn't, maybe you can ping someone from the diffusers team.

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.