- 24 Dec, 2022 18 commits
-
-
AUTOMATIC1111 authored
Fix various typos
-
AUTOMATIC1111 authored
Better support for 2.0-inpainting and 2.0-depth special models
-
AUTOMATIC authored
-
AUTOMATIC authored
-
AUTOMATIC1111 authored
Improve img2img color correction by performing a luminosity blend
-
AUTOMATIC1111 authored
Prevent overlapping in X/Y plot by changing font size
-
AUTOMATIC1111 authored
Add option to save upscaler to filename suffix in extras tab.
-
AUTOMATIC1111 authored
Fix docker tmp/ and extensions/ handling for docker.
-
AUTOMATIC1111 authored
Add upscale latent, VAE, styles as new axis options to X/Y plot
-
AUTOMATIC authored
add an option to samplers for skipping next to last sigma
-
AUTOMATIC1111 authored
Add a workaround patch for DPM2 a issue
-
AUTOMATIC1111 authored
Set httpcore version in requirements - fixes #4833
-
AUTOMATIC1111 authored
Training fixes for MPS
-
AUTOMATIC1111 authored
Improve InvokeAI cross attention reliability and speed when using MPS for large images
-
AUTOMATIC1111 authored
fix hints file typo
-
AUTOMATIC authored
-
AUTOMATIC1111 authored
fix: xformers
-
Akiba authored
-
- 21 Dec, 2022 1 commit
-
-
brkirch authored
Check if q.shape[0] * q.shape[1] is 2**18 or larger and use the lower memory usage MPS optimization if it is. This should prevent most crashes that were occurring at certain resolutions (e.g. 1024x1024, 2048x512, 512x2048). Also included is a change to check slice_size and prevent it from being divisible by 4096 which also results in a crash. Otherwise a crash can occur at 1024x512 or 512x1024 resolution.
-
- 20 Dec, 2022 1 commit
-
-
Florian Chrometz authored
-
- 19 Dec, 2022 1 commit
-
-
Alex "mcmonkey" Goodwin authored
-
- 18 Dec, 2022 1 commit
-
-
MMaker authored
Adds upscale latent space for hires., VAE, and Styles as new axis options to the X/Y plot.
-
- 17 Dec, 2022 6 commits
-
-
timntorres authored
-
timntorres authored
-
timntorres authored
-
brkirch authored
-
brkirch authored
When saving training results with torch.save(), an exception is thrown: "RuntimeError: Can't call numpy() on Tensor that requires grad. Use tensor.detach().numpy() instead." So for MPS, check if Tensor.requires_grad and detach() if necessary.
-
MMaker authored
This prevents scenarios where text without line breaks will start overlapping with each other when generating X/Y plots. This is most evident when generating X/Y plots with checkpoints, as most don't contain spaces and sometimes include extra information such as the epoch, making it extra long.
-
- 16 Dec, 2022 3 commits
-
-
Alex "mcmonkey" Goodwin authored
-
Alex "mcmonkey" Goodwin authored
DPM2 a and DPM2 a Karras samplers are both affected by an issue described by https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/3483 and can be resolved by a workaround suggested by the k-diffusion author at https://github.com/crowsonkb/k-diffusion/issues/43#issuecomment-1305195666
-
Akiba authored
-
- 15 Dec, 2022 1 commit
-
-
Jim Hays authored
-
- 11 Dec, 2022 6 commits
-
-
ThereforeGames authored
-
ThereforeGames authored
-
ThereforeGames authored
-
ThereforeGames authored
-
MrCheeze authored
unconditionally set use_ema=False if value not specified (True never worked, and all configs except v1-inpainting-inference.yaml already correctly set it to False)
-
Alex "mcmonkey" Goodwin authored
-
- 10 Dec, 2022 2 commits