Stuff to be fixed, in no particular order...
- Mysterious Extra network corruptions
- Unet type mismatch/autocast issues on some platforms
Stuff to be added, in no particular order...
- Diffusers:
- Add Hires
- Add Lora/Lyco mixer
- Add ControlNet
- Add SD and SD-XL Pix2Pix
- Fix DeepFloyd IF model
- Redo Prompt parser for diffusers
- Add unCLIP model
- Add ToMe support
- Add Training support
- Technical debt:
- Port A1111 stuff
- Port
p.all_hr_prompts
- Import core repos to reduce dependencies
- Update
gradio
- Parse StabilityAI
modelspec
metadata
- Non-technical:
- Create additional themes
- Update Wiki
- Get more high-quality upscalers
- Rename repo: automatic -> sdnext
- Localization
- New Minor
- Prompt padding for positive/negative
- XYZ grid upscalers
- Built-in
motd
-style notifications - Docker PR
- New Major
- Style editor (use json format instead of csv)
- Profile manager (for config.json and ui-config.json)
- Multi-user support
- Add SAG,(https://github.com/ashen-sensored/sd_webui_SAG)
- Image phash and hdash using
imagehash
- Model merge using
git-rebasin
- Enable refiner-style workflow for
ldm
backend - Add
sgm
backend - Cache models in VRAM
- Train:
- Use
interrogator
- Use
rembg
- Templates for SD-XL training
- Lora train UI
- Use
- Redesign
- Extensions reporting framework
- New UI
- New inpainting canvas controls (move from backend to purely frontend)
- New image browser (move from backend to purely frontend)
- New extra networks (move from backend to purely frontend)
- Change workflows from static/legacy to steps-based
Stuff to be investigated...
Pick & merge PRs from main repo...
- up-to-date with: df004be
- current todo list: https://github.com/AUTOMATIC1111/stable-diffusion-webui/compare/df004be...394ffa7
Tech that can be integrated as part of the core workflow...
- Git-ReBasin
- Null-text inversion
- Custom diffusion, Custom diffusion
- Dream artist
- QuickEmbedding
- DragGAN
- LamaCleaner
TensorRT
- Bunch of stuff: https://pharmapsychotic.com/tools.html