OPENGIT/stable-diffusion-webui
1
0
Fork 0
mirror of https://github.com/AUTOMATIC1111/stable-diffusion-webui.git synced 2025-02-07 06:02:53 +08:00
Code Issues Packages Projects Releases Wiki Activity

add auto's PR notes

ClashSAN 2023-07-27 15:10:20 -04:00
parent dffbab99b7
commit 5f09d4fca9
1 changed files with 14 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
Features.md

@ -5,12 +5,24 @@ All examples are non-cherrypicked unless specified otherwise.
# SD-XL
## SD-XL BASE
[PR](https://github.com/AUTOMATIC1111/stable-diffusion-webgenerative-models), [Github](https://github.com/Stability-AI/generative-models).
[PR](https://github.com/AUTOMATIC1111/stable-diffusion-webui/pull/11757), [Github](https://github.com/Stability-AI/generative-models).
### Info
This is a txt2img model designed for generating quality 1024×1024-sized images. It is not meant to generate good pictures at 512×512.
It's tested to produce same (or very close) images as Stability-AI's repo (need to set Random number generator source = CPU in settings)
- textual inversion should not work, embeddings need to be created specifically for SDXL.
- train tab will not work.
- DDIM, PLMS, UniPC samplers do not work for SDXL
- `--lowvram`, `--medvram` works
- attention optimizations work
- SDXL Loras work
- works at minimum 4gb gpu (30XX) (use --lowvram --xformers for max of 1280×1280)
>NOTE: Initial loading of these models require those with 24gb cpu memory (RAM) and under to setup a large pagefile. This may change in the future.
@ -23,6 +35,7 @@ This is a txt2img model designed for generating quality 1024×1024-sized images.
(Please use both of these to avoid performance issues.)
# SD2 Variation Models
[PR](https://github.com/AUTOMATIC1111/stable-diffusion-webui/pull/8958), ([more info.](https://github.com/Stability-AI/stablediffusion/blob/main/doc/UNCLIP.MD))