Mirror/stable-diffusion-webui
1
0
Fork 0
mirror of https://github.com/AUTOMATIC1111/stable-diffusion-webui.git synced 2024-06-07 21:20:49 +00:00
Code Issues Packages Projects Releases Wiki Activity
6,403 Commits 35 Branches 33 Tags 305 MiB
976c1053ef
Commit Graph

514 Commits

Author SHA1 Message Date
AUTOMATIC1111
d73db17ee3
Merge pull request #12515 from catboxanon/fix/gc1 
Clear sampler and garbage collect before decoding images to reduce VRAM
 2023-08-13 16:45:38 +03:00
AUTOMATIC1111
d8419762c1 Lora: output warnings in UI rather than fail for unfitting loras; switch to logging for error output in console 2023-08-13 15:07:37 +03:00
catboxanon
69f49c8d39 Clear sampler before decoding images 
More significant VRAM reduction.
 2023-08-13 04:40:34 -04:00
AUTOMATIC1111
599f61a1e0 use dataclass for StableDiffusionProcessing 2023-08-13 08:24:16 +03:00
AUTOMATIC1111
fa9370b741 add refiner to StableDiffusionProcessing class 
write out correct model name in infotext, rather than the refiner model
 2023-08-13 06:07:30 +03:00
AUTOMATIC1111
b293ed3061 make it possible to use hires fix together with refiner 2023-08-12 12:54:32 +03:00
AUTOMATIC1111
64311faa68 put refiner into main UI, into the new accordions section 
add VAE from main model into infotext, not from refiner model
option to make scripts UI without gr.Group
fix inconsistencies with refiner when usings samplers that do more denoising than steps
 2023-08-12 12:39:59 +03:00
catboxanon
af27b716e5 Fix color correction by converting image to RGB 2023-08-11 12:22:11 -04:00
AUTOMATIC1111
ae6b30907d
Merge pull request #12470 from Splendide-Imaginarius/mask-blur-property+kernel 
Make `StableDiffusionProcessingImg2Img.mask_blur` a property, make more inline with PIL `GaussianBlur`
 2023-08-11 15:03:18 +03:00
Robert Barron
863613293e Merge branch 'shared-hires-prompt-raw' into shared-hires-prompt-test 2023-08-10 07:45:35 -07:00
AUTOMATIC1111
70a01cd444 Merge branch 'dev' into refiner 2023-08-10 17:04:38 +03:00
AUTOMATIC1111
9d78d317ae add VAE to infotext 2023-08-10 16:22:10 +03:00
AUTOMATIC1111
1b3093fe3a fix --use-textbox-seed 2023-08-10 15:58:53 +03:00
AUTOMATIC1111
33446acf47 face restoration and tiling moved to settings - use "Options in main UI" setting if you want them back 2023-08-10 12:41:41 +03:00
AUTOMATIC1111
9199b6b7eb add a custom UI element that combines accordion and checkbox 
rework hires fix UI to use accordion
prevent bogus progress output in console when calculating hires fix dimensions
 2023-08-10 11:20:46 +03:00
catboxanon
259805947e
Add slerp import for extension backwards compat 2023-08-09 14:24:16 -04:00
Robert Barron
d1ba46b6e1 allow first pass and hires pass to use a single prompt to do different prompt editing, hires is 1.0..2.0: 
relative time range is [1..2]
  absolute time range is [steps+1..steps+hire_steps], e.g. with 30 steps and 20 hires steps, '20' is 2/3rds through first pass, and 40 is halfway through hires pass
 2023-08-09 10:38:47 -07:00
AUTOMATIC1111
0d5dc9a6e7 rework RNG to use generators instead of generating noises beforehand 2023-08-09 08:43:31 +03:00
AUTOMATIC1111
d81d3fa8cd fix styles missing from the prompt in infotext when making a grid of batch of multiplie images 2023-08-09 07:45:06 +03:00
AUTOMATIC1111
a2360de3f3
Merge pull request #12412 from dhwz/dev 
fix typo
 2023-08-08 23:30:57 +03:00
AUTOMATIC1111
ec194b6374 fix webui not switching back to original model from refiner when batch count is greater than 1 2023-08-08 22:14:02 +03:00
AUTOMATIC1111
54c3e5c913 Merge branch 'dev' into refiner 2023-08-08 21:49:47 +03:00
AUTOMATIC1111
a8a256f9b5 REMOVE 2023-08-08 21:08:50 +03:00
dhwz
2a72d76d6f fix typo 2023-08-08 19:08:37 +02:00
AUTOMATIC1111
5a0db84b6c add infotext 
add proper support for recalculating conds in k-diffusion samplers
remove support for compvis samplers
 2023-08-06 17:53:33 +03:00
AUTOMATIC1111
f1975b0213 initial refiner support 2023-08-06 17:01:07 +03:00
AUTOMATIC1111
57e8a11d17 enable cond cache by default 2023-08-06 13:25:51 +03:00
AUTOMATIC1111
5cae08f2c3 fix rework saving incomplete images 2023-08-06 06:55:19 +03:00
AUTOMATIC1111
ee96a6a588 do the same for s_tmax #12345 2023-08-06 06:32:41 +03:00
AUTOMATIC1111
ee75416e3e
Merge branch 'dev' into fix/s-noise 2023-08-06 06:25:35 +03:00
AUTOMATIC1111
d86d12e911 rework saving incomplete images 2023-08-06 06:21:36 +03:00
AUTOMATIC1111
2844d9597b
Merge pull request #12338 from AUTOMATIC1111/dont-save-incomplete-images 
don't save incomplete images
 2023-08-06 06:05:47 +03:00
catboxanon
f18a032190 Correct s_noise fix 2023-08-05 23:05:25 -04:00
catboxanon
85c2c138d2 Attempt to read s_tmax from arg first if option not found 2023-08-05 21:51:46 -04:00
catboxanon
c11104fed5 Add s_tmax 2023-08-05 21:42:03 -04:00
catboxanon
496cef956b Allow s_noise override to actually be used 2023-08-05 21:14:13 -04:00
w-e-w
1d7dcdb6c3 Option to not save incomplete images 2023-08-05 19:07:53 +09:00
AUTOMATIC1111
3f451f3042 do not add VAE Encoder/Decoder to infotext if it's the default 2023-08-05 10:36:26 +03:00
Kohaku-Blueleaf
b85ec2b9b6 Fix some merge mistakes 2023-08-05 13:14:00 +08:00
Kohaku-Blueleaf
d56a9cfe6a Merge branch 'dev' into efficient-vae-methods 2023-08-05 13:12:37 +08:00
Splendide Imaginarius
56888644a6 Reduce mask blur kernel size to 2.5 sigmas 
This more closely matches the old behavior of PIL's Gaussian blur, and
fixes breakage when tiling.

See https://github.com/Coyote-A/ultimate-upscale-for-automatic1111/issues/111#issuecomment-1663504109

Thanks to Алексей Трофимов and eunnone for reporting the issue.
 2023-08-05 04:54:23 +00:00
AUTOMATIC1111
0ae2767ae6
Merge pull request #12181 from AUTOMATIC1111/hires_checkpoint 
Hires fix change checkpoint
 2023-08-05 07:47:34 +03:00
Kohaku-Blueleaf
aa744cadc8 add infotext 2023-08-05 12:35:40 +08:00
Kohaku-Blueleaf
21000f13a1 replace get_first_stage_encoding 2023-08-04 18:23:14 +08:00
Kohaku-Blueleaf
6346d8eeaa Revert "change all encode" 
This reverts commit 094c416a80.
 2023-08-04 17:53:30 +08:00
Kohaku-Blueleaf
094c416a80 change all encode 2023-08-04 17:53:16 +08:00
Kohaku-Blueleaf
70e66e81e5 Merge branch 'dev' into efficient-vae-methods 2023-08-04 14:38:16 +08:00
AUTOMATIC1111
f0c1063a70 resolve some of circular import issues for kohaku 2023-08-04 09:13:46 +03:00
Kohaku-Blueleaf
75336dfc84 add TAESD for i2i and t2i 2023-08-04 13:38:52 +08:00
Splendide Imaginarius
a1825ee741 Make StableDiffusionProcessingImg2Img.mask_blur a property 
Fixes breakage when mask_blur is set after construction.

See https://github.com/Coyote-A/ultimate-upscale-for-automatic1111/issues/111#issuecomment-1652091424

Thanks to Алексей Трофимов and eunnone for reporting the issue.
 2023-08-03 02:07:00 +00:00
AUTOMATIC1111
84b6fcd02c add NV option for Random number generator source setting, which allows to generate same pictures on CPU/AMD/Mac as on NVidia videocards. 2023-08-03 00:00:23 +03:00
AUTOMATIC1111
ccb9233934 add yet another torch_gc to reclaim some of VRAM after the initial stage of img2img 2023-08-02 18:53:09 +03:00
AUTOMATIC1111
c09bc2c608 fix "clamp_scalar_cpu" not implemented for 'Half' 2023-07-31 13:20:26 +03:00
AUTOMATIC1111
29d7e31d89 repair AttributeError: 'NoneType' object has no attribute 'conditioning_key' 2023-07-31 10:43:26 +03:00
AUTOMATIC1111
dca121e903 set the field to None instead 2023-07-31 09:13:07 +03:00
AUTOMATIC1111
0af4127fd1 delete the field that is preventing the model from being unloaded and is causing increased RAM usage 2023-07-30 19:36:24 +03:00
AUTOMATIC1111
02038036ff make it so that VAE NaNs autodetection also works during first pass of hires fix 2023-07-30 16:16:31 +03:00
AUTOMATIC1111
cc53db6652 this time for sure 2023-07-30 15:30:33 +03:00
AUTOMATIC1111
a64fbe8928 make it possible to use checkpoints of different types (SD1, SDXL) in first and second pass of hires fix 2023-07-30 15:12:09 +03:00
AUTOMATIC1111
eec540b227 repair non-latent upscaling broken for SDXL 2023-07-30 15:04:12 +03:00
AUTOMATIC1111
77761e7bad linter 2023-07-30 14:10:33 +03:00
AUTOMATIC1111
40cd59207b make it work with SDXL 2023-07-30 14:10:26 +03:00
AUTOMATIC1111
3bca90b249 hires fix checkpoint selection 2023-07-30 13:48:27 +03:00
AUTOMATIC1111
835a7dbf0e simplify PostprocessBatchListArgs 2023-07-26 07:49:57 +03:00
AUTOMATIC1111
7c22bbd3ad attempt 2 2023-07-26 07:04:07 +03:00
AUTOMATIC1111
ae36e0899f alternative solution for infotext issue 2023-07-26 06:36:06 +03:00
ljleb
5b06607476 simplify 2023-07-24 15:43:06 -04:00
ljleb
6b68b59032 use local vars 2023-07-24 15:38:52 -04:00
ljleb
ca45ff1ae6 add postprocess_batch_list callback 2023-07-24 13:52:24 -04:00
AUTOMATIC1111
23c947ab03 automatically switch to 32-bit float VAE if the generated picture has NaNs. 2023-07-19 20:23:30 +03:00
AUTOMATIC1111
4334d25978 bugfix: model name was added together with directory name to infotext and to [model_name] filename pattern 2023-07-19 15:49:54 +03:00
w-e-w
c8b55f29e2 missing p save_image before-highres-fix 2023-07-19 08:27:19 +09:00
AUTOMATIC1111
570f42afd1 possible fix for FP16 VAE failing in img2img SDXL 2023-07-16 12:28:50 +03:00
AUTOMATIC1111
0198eaec45
Merge pull request #11757 from AUTOMATIC1111/sdxl 
SD XL support
 2023-07-16 12:04:53 +03:00
AUTOMATIC1111
2b1bae0d75 add textual inversion hashes to infotext 2023-07-15 08:41:22 +03:00
AUTOMATIC1111
14cf434bc3 fix an issue in live previews that happens when you use SDXL with fp16 VAE 2023-07-15 07:33:16 +03:00
AUTOMATIC1111
471a5a66b7 add more relevant fields to caching conds 2023-07-14 17:54:09 +03:00
AUTOMATIC1111
594c8e7b26 fix CLIP doing the unneeded normalization 
revert SD2.1 back to use the original repo
add SDXL's force_zero_embeddings to negative prompt
 2023-07-13 11:35:52 +03:00
AUTOMATIC1111
da464a3fb3 SDXL support 2023-07-12 23:52:43 +03:00
Aarni Koskela
da468a585b Fix typo: checkpoint_alisases 2023-07-08 17:28:42 +03:00
AUTOMATIC1111
53924aeaf0
Merge pull request #11477 from hako-mikan/master 
add `before_hr` script callback
 2023-07-08 14:43:06 +03:00
AUTOMATIC1111
9be8903ca9
Merge pull request #11567 from AUTOMATIC1111/seed_resize_to_0 
Don't add "Seed Resize: -1x-1" to API image metadata
 2023-07-08 13:58:31 +03:00
missionfloyd
f731a728c6
Check seed_resize_from <= 0 2023-07-03 11:41:10 -06:00
missionfloyd
7f46f81dd7
Change default seed_resize to 0 2023-07-01 17:20:56 -06:00
hako-mikan
b0ec69b360
add 'before_hr callback' script callback 2023-06-28 18:37:08 +09:00
AUTOMATIC1111
fab73f2e7d
Merge pull request #11325 from stablegeniusdiffuser/dev-batch-grid-metadata 
Add parameter to differentiate between batch run grids or ordinary images to write proper metadata
 2023-06-27 14:23:39 +03:00
AUTOMATIC1111
4147fd6b2f
Merge branch 'dev' into 10141-gradio-user-exif 2023-06-27 09:05:53 +03:00
stablegeniusdiffuser
27e9e3f6fa Add use_main_prompt parameter to use proper metadata for batch run grids or individual images 2023-06-19 20:36:44 +02:00
Jared Deckard
f603275d84 Add an opt-in infotext user name setting 2023-06-15 11:00:20 -05:00
Jared Deckard
d3c86e5178 Note the Gradio user in the Exif data 2023-06-14 17:15:52 -05:00
AUTOMATIC
cfdd1b9418 linter 2023-06-09 22:47:27 +03:00
AUTOMATIC1111
d00139eea8
Merge pull request #11087 from AUTOMATIC1111/persistent_conds_cache 
persistent conds cache
 2023-06-09 22:32:49 +03:00
Splendide Imaginarius
1503af60b0 Split mask blur into X and Y components 
Prequisite to fixing Outpainting MK2 mask blur bug.
 2023-06-09 08:36:33 +00:00
w-e-w
7f2214aa2b persistent conds cache 
Update shared.py
 2023-06-08 14:27:22 +09:00
AUTOMATIC1111
7a7a201d81
Merge pull request #10956 from akx/len 
Simplify a bunch of `len(x) > 0`/`len(x) == 0` style expressions
 2023-06-05 11:06:37 +03:00
AUTOMATIC
fbf88343de prevent calculating cons for second pass of hires fix when they are the same as for the first pass 2023-06-04 16:29:02 +03:00
AUTOMATIC
1ca5e76f7b fix for conds of second hires fox pass being calculated using first pass's networks, and add an option to revert to old behavior 2023-06-04 13:07:31 +03:00
w-e-w
1c9d1b0ee0 simplify self.extra_network_data 2023-06-04 05:19:34 +09:00
w-e-w
f098e726d3 fix conds caching with extra network 2023-06-04 04:24:44 +09:00
Aarni Koskela
51864790fd Simplify a bunch of len(x) > 0/len(x) == 0 style expressions 2023-06-02 15:07:10 +03:00