huggingface / huggingface/diffusion-models-class
How can I use stable-diffusion-2-1?
Open
- Dominant language
- Jupyter Notebook
- Stars
- 4.4k
- Forks
- 492
- PR merge metrics
- No merged PRs in 30d
Description
when I use stable-diffusion-2-1(768ema), something wrong happened. It seems that 768ema can't correctly sample noise and inverse image. but 2-1-base works well. Is there any differences between them? How can I use stable-diffusion-2-1(768ema)? the cause I studying is unit4, 01_ddim_inversion.ipynb.
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.