michael/stable-diffusion-webui
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
stable-diffusion-webui/modules
History
brkirch e3b53fd295 Add UI setting for upcasting attention to float32 
Adds "Upcast cross attention layer to float32" option in Stable Diffusion settings. This allows for generating images using SD 2.1 models without --no-half or xFormers.

In order to make upcasting cross attention layer optimizations possible it is necessary to indent several sections of code in sd_hijack_optimizations.py so that a context manager can be used to disable autocast. Also, even though Stable Diffusion (and Diffusers) only upcast q and k, unfortunately my findings were that most of the cross attention layer optimizations could not function unless v is upcast also.
2023-01-25 01:13:04 -05:00
..
api also return the removed field to sdapi/v1/upscalers because someone might have relied on it existing 2023-01-24 10:09:30 +03:00
codeformer fix F541 f-string without any placeholders 2022-12-24 21:35:29 +02:00
hypernetworks enable compact view for train tab 2023-01-22 00:02:51 +03:00
textual_inversion extra networks UI 2023-01-21 08:36:07 +03:00
call_queue.py big rework of progressbar/preview system to allow multiple users to prompts at the same time and do not get previews of each other 2023-01-15 18:51:04 +03:00
codeformer_model.py Set device for facelib/facexlib and gfpgan 2022-11-12 03:34:13 -05:00
deepbooru.py added an option to filter out deepbooru tags 2022-12-24 16:22:47 +03:00
deepbooru_model.py Add option for float32 sampling with float16 UNet 2023-01-25 01:13:02 -05:00
devices.py Add UI setting for upcasting attention to float32 2023-01-25 01:13:04 -05:00
errors.py better support for xformers flash attention on older versions of torch 2023-01-23 16:40:20 +03:00
esrgan_model.py Revert "MPS Upscalers Fix" 2022-11-17 00:08:21 -05:00
esrgan_model_arch.py fix typo in filename for ESRGAN arch 2022-10-09 15:08:27 +03:00
extensions.py add built-in extension system 2022-12-03 18:06:33 +03:00
extra_networks.py extra networks UI 2023-01-21 08:36:07 +03:00
extra_networks_hypernet.py Lora support! 2023-01-21 16:15:53 +03:00
extras.py third time's the charm 2023-01-23 14:50:20 +03:00
face_restoration.py codeformer support 2022-09-07 12:32:28 +03:00
generation_parameters_copypaste.py extra networks UI 2023-01-21 08:36:07 +03:00
gfpgan_model.py Set device for facelib/facexlib and gfpgan 2022-11-12 03:34:13 -05:00
hashes.py increase block size 2023-01-15 09:42:34 -05:00
images.py make it so that PNG images with EXIF do not lose parameters in PNG info tab 2023-01-17 23:54:23 +03:00
img2img.py bump gradio to 3.16.2 2023-01-18 13:59:45 +03:00
import_hook.py add a comment for disable xformers hack 2022-12-24 07:57:56 +03:00
interrogate.py remove fairscale requirement, add fake fairscale to make BLIP not complain about it mk2 2023-01-24 20:30:43 +03:00
localization.py add ability to create extensions that add localizations 2022-11-06 09:02:35 +03:00
lowvram.py unload depth model if medvram/lowvram enabled 2022-12-10 11:02:47 -05:00
masking.py Fix non-square full resolution inpainting. 2022-10-30 15:33:02 -07:00
memmon.py add additional memory states 2022-12-30 19:36:36 -05:00
modelloader.py add more stuff to ignore when creating model from config 2023-01-10 16:51:04 +03:00
ngrok.py Correct singleton comparisons 2022-12-14 20:59:33 +02:00
paths.py fix BLIP failing to import depending on configuration 2023-01-24 00:24:17 +03:00
postprocessing.py a possible fix for broken image upscaling 2023-01-23 17:59:58 +03:00
processing.py Add UI setting for upcasting attention to float32 2023-01-25 01:13:04 -05:00
progress.py add progress bar to modelmerger 2023-01-19 09:25:37 +03:00
prompt_parser.py add BREAK keyword to end current text chunk and start the next 2023-01-15 22:29:53 +03:00
realesrgan_model.py Fix unable to find Real-ESRGAN model info error (AttributeError: 'NoneType' object has no attribute 'data_path') #6841 #5170 2023-01-17 13:57:55 +03:00
safe.py Merge pull request #5992 from yuvalabou/F541 2022-12-25 09:16:08 +03:00
script_callbacks.py Lora support! 2023-01-21 16:15:53 +03:00
script_loading.py make existing script loading and new preload code use same code for loading modules 2022-11-12 10:56:06 +03:00
scripts.py rework extras tab to use script system 2023-01-23 09:24:43 +03:00
scripts_postprocessing.py rework extras tab to use script system 2023-01-23 09:24:43 +03:00
sd_disable_initialization.py fix missing field for aesthetic embedding extension 2023-01-22 08:20:48 +03:00
sd_hijack.py write a comment for fix_checkpoint function 2023-01-19 20:39:10 +03:00
sd_hijack_checkpoint.py add option to show/hide warnings 2023-01-18 23:04:24 +03:00
sd_hijack_clip.py add BREAK keyword to end current text chunk and start the next 2023-01-15 22:29:53 +03:00
sd_hijack_clip_old.py CLIP hijack rework 2023-01-07 01:46:13 +03:00
sd_hijack_inpainting.py find configs for models at runtime rather than when starting 2023-01-04 12:47:42 +03:00
sd_hijack_open_clip.py Add support Stable Diffusion 2.0 2022-11-26 16:10:46 +03:00
sd_hijack_optimizations.py Add UI setting for upcasting attention to float32 2023-01-25 01:13:04 -05:00
sd_hijack_unet.py Add option for float32 sampling with float16 UNet 2023-01-25 01:13:02 -05:00
sd_hijack_utils.py Add option for float32 sampling with float16 UNet 2023-01-25 01:13:02 -05:00
sd_hijack_xlmr.py alt-diffusion integration 2022-12-31 18:06:35 +03:00
sd_models.py Add option for float32 sampling with float16 UNet 2023-01-25 01:13:02 -05:00
sd_samplers.py Add a check and explanation for tensor with all NaNs. 2023-01-16 22:59:46 +03:00
sd_vae.py allow baking in VAE in checkpoint merger tab 2023-01-19 10:39:51 +03:00
sd_vae_approx.py Fix Approx NN on devices other than CUDA 2023-01-15 08:13:33 -05:00
shared.py Add UI setting for upcasting attention to float32 2023-01-25 01:13:04 -05:00
styles.py change style dropdowns to multiselect 2023-01-14 14:56:39 +03:00
sub_quadratic_attention.py Add UI setting for upcasting attention to float32 2023-01-25 01:13:04 -05:00
txt2img.py bump gradio to 3.16.2 2023-01-18 13:59:45 +03:00
ui.py fix open directory button failing 2023-01-23 15:50:32 +03:00
ui_common.py fix open directory button failing 2023-01-23 15:50:32 +03:00
ui_components.py rework extras tab to use script system 2023-01-23 09:24:43 +03:00
ui_extensions.py sort extensions by date and add an option to sort by other columns 2023-01-06 12:32:44 +03:00
ui_extra_networks.py add support for apostrophe in extra network names 2023-01-23 23:11:34 +03:00
ui_extra_networks_hypernets.py add a slider for default value of added extra networks 2023-01-22 00:02:52 +03:00
ui_extra_networks_textual_inversion.py fix broken textual inversion extras tab 2023-01-22 00:16:26 +03:00
ui_postprocessing.py rework extras tab to use script system 2023-01-23 09:24:43 +03:00
ui_tempdir.py add support for running with gradio 3.9 installed 2023-01-03 14:18:48 +03:00
upscaler.py Fix unable to find Real-ESRGAN model info error (AttributeError: 'NoneType' object has no attribute 'data_path') #6841 #5170 2023-01-17 13:57:55 +03:00
xlmr.py fix bugs 2022-11-30 14:56:12 +08:00