Age | Commit message (Collapse) | Author |
|
|
|
|
|
|
|
(Optimization) Option to remove negative conditioning at low sigma values
|
|
Fix prompt schedule for second order samplers
|
|
|
|
|
|
add tiled inference support for ScuNET
|
|
copypaste support to RNG source
|
|
Option to make images generated from a given manual seed consistent across CUDA and MPS devices
|
|
Automatically select current word when adjusting weight with ctrl+up/down
|
|
Add support for saving init images in img2img
|
|
|
|
|
|
Makes a given manual seed generate the same images across different
platforms, independently of the GPU architecture in use.
Fixes #9613.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
feat: try sort as ignore-case
|
|
|
|
make launch.py use the same command line argument parser as the main program
|
|
|
|
|
|
|
|
Add ability to display and/or save inpainting mask and masked composite
|
|
more stylistic changes
|
|
|
|
For inpainting, this exposes the mask and masked composite and gives
the user the ability to display these in the web UI,
save to disk, or both.
|
|
|
|
https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/8368
|
|
|
|
adds checks for resulting image size to avoid memory issues
|
|
allow configurable hiding of ui tabs
|
|
Add lossless webp option
|
|
|
|
Add scaled dot product attention
|
|
|
|
|
|
|
|
|
|
|
|
|