Commit Graph

352 Commits

Author SHA1 Message Date
lllyasviel
f06ba8e60b Significantly reduce thread abuse for faster model moving
This will move all major gradio calls into the main thread rather than random gradio threads.
This ensures that all torch.module.to() are performed in main thread to completely possible avoid GPU fragments.
In my test now model moving is 0.7 ~ 1.2 seconds faster, which means all 6GB/8GB VRAM users will get 0.7 ~ 1.2 seconds faster per image on SDXL.
2024-02-08 10:13:59 -08:00
lllyasviel
66187d97fb early initilize 2024-01-31 15:27:13 -08:00
AUTOMATIC1111
f0e2e8b930 update #14354 2023-12-30 15:12:48 +03:00
Muhammad Rehan Aslam
0d5941edbc
Update webui.py
Co-authored-by: Aarni Koskela <akx@iki.fi>
2023-12-19 09:50:38 +05:00
Muhammad Rehan Aslam
fe4d084390
Update webui.py
Added (Fixed) IPV6 Functionality When there is No Webui Argument Passed
2023-12-18 17:50:00 +05:00
w-e-w
18e3e6d6ab consolidate local check 2023-08-28 03:43:27 +09:00
w-e-w
e422f19ee9 non-local condition 2023-08-28 03:27:07 +09:00
AUTOMATIC1111
edfae9e78a add --loglevel commandline argument for logging
remove the progressbar for extension installation in favor of logging output
2023-08-09 20:49:33 +03:00
AUTOMATIC1111
95821f0132 split webui.py's initialization and utility functions into separate files 2023-08-09 18:11:13 +03:00
AUTOMATIC1111
386245a264 split shared.py into multiple files; should resolve all circular reference import errors related to shared.py 2023-08-09 10:25:35 +03:00
w-e-w
f17c8c2eff
Merge branch 'dev' into auro-autolaunch 2023-08-08 11:39:34 +09:00
w-e-w
c75bda867b setting: Automatically open webui in browser on startup 2023-08-08 11:29:33 +09:00
王怀宗
250a95b6fe fix: enable before_ui_callback when api only mode (fixes #7984) 2023-08-07 18:08:07 +08:00
AUTOMATIC1111
c96e4750d8 SD VAE rework 2
- the setting for preferring opts.sd_vae has been inverted and reworded
- resolve_vae function made easier to read and now returns an object rather than a tuple
- if the checkbox for overriding per-model preferences is checked, opts.sd_vae overrides checkpoint user metadata
- changing VAE in user metadata  for currently loaded model immediately applies the selection
2023-08-07 08:07:20 +03:00
AUTOMATIC1111
073c0ebba3 add gradio version warning 2023-08-04 08:04:23 +03:00
AUTOMATIC1111
af528552d6 fix linter issues 2023-08-03 23:31:14 +03:00
AUTOMATIC1111
ac81c1dd1f
Merge pull request #11958 from AUTOMATIC1111/conserve-ram
Use less RAM when creating models
2023-07-29 07:43:04 +03:00
w-e-w
b1a16a298c api only subpath (rootpath)
Co-Authored-By: 陈杰 <pythias@gmail.com>
2023-07-25 20:51:27 +09:00
AUTOMATIC1111
0a89cd1a58 Use less RAM when creating models 2023-07-24 22:08:08 +03:00
Jabasukuriputo Wang
118529a6dc
typo fix 2023-07-21 21:49:33 +08:00
AUTOMATIC1111
a99d5708e6 skip installing packages with pip if theyare already installed
record time it took to launch
2023-07-17 20:10:24 +03:00
AUTOMATIC1111
18256c5f01 fix for #11478 2023-07-08 14:58:33 +03:00
AUTOMATIC1111
211c3398f6
Merge pull request #11478 from catalpaaa/subpath
Fixing --subpath on newer gradio version
2023-07-08 14:53:42 +03:00
Aarni Koskela
b70001e618 Add SD_WEBUI_LOG_LEVEL envvar 2023-07-03 11:03:41 +03:00
catalpaaa
24d4475bdb fixing --subpath on newer gradio version 2023-06-28 03:15:01 -07:00
w-e-w
d80962681a remove fastapi.Response 2023-06-12 18:21:01 +09:00
w-e-w
b9664ab615 move _stop route to api 2023-06-12 18:15:27 +09:00
w-e-w
c2808f3040 SD_WEBUI_RESTARTING 2023-06-06 02:52:05 +09:00
w-e-w
eaace155ce restore old disable --autolaunch 2023-06-06 02:47:18 +09:00
w-e-w
1411a6e74b rework-disable-autolaunch 2023-06-06 01:09:30 +09:00
w-e-w
8f8405274c remove redundant 2023-06-02 17:18:42 +09:00
w-e-w
2bbe3f5f0a remove redundant call list_optimizers() 2023-06-02 16:51:15 +09:00
AUTOMATIC
b3390a9840 Merge branch 'dev' into startup-profile 2023-06-01 08:42:50 +03:00
Sakura-Luna
cf07983a6e
Upgrade xformers 2023-05-28 20:42:19 +08:00
AUTOMATIC
339b531570 custom unet support 2023-05-27 15:47:33 +03:00
AUTOMATIC
a6e653be26 possible fix for empty list of optimizations #10605 2023-05-23 18:49:15 +03:00
AUTOMATIC1111
809001fe41
Merge pull request #10623 from akx/bump-gradio
Bump gradio to 3.32
2023-05-22 22:18:05 +03:00
Aarni Koskela
47b669bc9f Upgrade Gradio, remove docs URL hack 2023-05-22 09:53:24 +03:00
AUTOMATIC
8faac8b963 run basic torch calculation at startup in parallel to reduce the performance impact of first generation 2023-05-21 21:55:14 +03:00
AUTOMATIC
0cc05fc492 work on startup profile display 2023-05-21 00:41:41 +03:00
Aarni Koskela
df004be2fc Add a couple from __future__ import annotationses for Py3.9 compat 2023-05-21 00:26:16 +03:00
AUTOMATIC
05e6fc9aa9 Merge branch 'ui-selection-for-cross-attention-optimization' into dev 2023-05-20 22:29:51 +03:00
Aarni Koskela
71f4a4afdf Deduplicate webui.py initial-load/reload code 2023-05-19 17:38:42 +03:00
Aarni Koskela
0f28aee9cd Refactor gradio auth 2023-05-19 17:35:51 +03:00
Aarni Koskela
674e80c625 Note pending PR for app_kwargs 2023-05-19 17:35:51 +03:00
Aarni Koskela
8a178e6717 Refactor configure opts_onchange out 2023-05-19 17:35:51 +03:00
Aarni Koskela
8200e0c27b Refactor configure_sigint_handler out 2023-05-19 17:35:51 +03:00
Aarni Koskela
1482c89376 Refactor validate_tls_options out, fix typo (keyfile was there twice) 2023-05-19 17:35:51 +03:00
Aarni Koskela
a0005121ae Simplify CORS middleware configuration 2023-05-19 15:37:13 +03:00
Aarni Koskela
21ee46eea7 Deduplicate default extra network registration 2023-05-19 15:35:16 +03:00