diffusers does not support AutoencoderKL of SD 2.0
#15
by
Chalet37
- opened
I use dreambooth in https://github.com/huggingface/diffusers/tree/main/examples/dreambooth, and I found it could not load vae model correctly. It seems https://huggingface.co/stabilityai/stable-diffusion-2-base/blob/main/vae/diffusion_pytorch_model.bin was trained from different AutoencoderKL.
How to use sd 2.0base in dreambooth?
Chalet37
changed discussion status to
closed