search

Bing-su / adetailer

Auto detecting, masking and inpainting with detection model.
GNU Affero General Public License v3.0
4.12k stars 318 forks source link

[Bug]: Brown square over faces. #370

Closed Fot3k closed 10 months ago

Fot3k commented 11 months ago

Describe the bug

When enabling ADetailer it keeps generating a brown square over faces. When disabled it generates fine.

Issue is the same on all Models/VAE/Loras etc.

Using AMD with Vlad SD.Next.

I managed to fix it once by messing with the SD settings but I can't pinpoint what is causing it.

Screenshots

image

Console logs, from start to end.

PS E:\AI Software\automatic> .\webui.bat --use-directml --autolaunch --debug --upgrade
Using VENV: E:\AI Software\automatic\venv
11:33:54-005380 DEBUG    Logger: file=E:\AI Software\automatic\sdnext.log level=10 size=0 mode=create
11:33:54-008389 INFO     Starting SD.Next
11:33:54-013544 INFO     Python 3.10.11 on Windows
11:33:54-128372 INFO     Version: app=sd.next updated=2023-10-17 hash=379fe1f3 url=https://github.com/vladmandic/automatic.git/tree/master
11:33:54-498701 INFO     Platform: arch=AMD64 cpu=AMD64 Family 25 Model 33 Stepping 0, AuthenticAMD system=Windows release=Windows-10-10.0.22621-SP0 python=3.10.11
11:33:54-500701 DEBUG    Setting environment tuning
11:33:54-501701 DEBUG    Torch overrides: cuda=False rocm=False ipex=False diml=True openvino=False
11:33:54-502703 DEBUG    Torch allowed: cuda=False rocm=False ipex=False diml=True openvino=False
11:33:54-503701 INFO     Using DirectML Backend
11:33:54-504701 INFO     Installing torch - this may take a while
11:33:54-565534 INFO     Verifying requirements
11:33:54-576873 INFO     Verifying packages
11:33:54-577874 INFO     Verifying repositories
11:33:54-632006 DEBUG    Submodule: E:\AI Software\automatic\repositories\stable-diffusion-stability-ai / main
11:33:55-357078 DEBUG    Submodule: E:\AI Software\automatic\repositories\taming-transformers / master
11:33:57-214356 DEBUG    Submodule: E:\AI Software\automatic\repositories\BLIP / main
11:33:57-865212 INFO     Verifying submodules
11:33:59-864913 DEBUG    Submodule: extensions-builtin/clip-interrogator-ext / main
11:34:00-503715 DEBUG    Submodule: extensions-builtin/sd-extension-chainner / main
11:34:01-289623 DEBUG    Submodule: extensions-builtin/sd-extension-system-info / main
11:34:01-946895 DEBUG    Submodule: extensions-builtin/sd-webui-agent-scheduler / main
11:34:02-867704 DEBUG    Submodule: extensions-builtin/sd-webui-controlnet / main
11:34:03-532380 DEBUG    Submodule: extensions-builtin/stable-diffusion-webui-images-browser / main
11:34:04-188415 DEBUG    Submodule: extensions-builtin/stable-diffusion-webui-rembg / master
11:34:04-856417 DEBUG    Submodule: modules/lora / main
11:34:05-507972 DEBUG    Submodule: wiki / master
11:34:06-286710 DEBUG    Installed packages: 258
11:34:06-287710 DEBUG    Extensions all: ['a1111-sd-webui-lycoris', 'clip-interrogator-ext', 'LDSR', 'Lora', 'multidiffusion-upscaler-for-automatic1111', 'ScuNET', 'sd-extension-chainner',
                         'sd-extension-system-info', 'sd-webui-agent-scheduler', 'sd-webui-controlnet', 'stable-diffusion-webui-images-browser', 'stable-diffusion-webui-rembg', 'SwinIR']
11:34:06-338224 DEBUG    Submodule: extensions-builtin\a1111-sd-webui-lycoris / main
11:34:07-099386 DEBUG    Submodule: extensions-builtin\clip-interrogator-ext / main
11:34:07-741940 DEBUG    Running extension installer: E:\AI Software\automatic\extensions-builtin\clip-interrogator-ext\install.py
11:34:14-038122 DEBUG    Submodule: extensions-builtin\multidiffusion-upscaler-for-automatic1111 / main
11:34:15-467125 DEBUG    Submodule: extensions-builtin\sd-extension-chainner / main
11:34:16-203228 DEBUG    Submodule: extensions-builtin\sd-extension-system-info / main
11:34:16-797110 DEBUG    Running extension installer: E:\AI Software\automatic\extensions-builtin\sd-extension-system-info\install.py
11:34:17-202581 DEBUG    Submodule: extensions-builtin\sd-webui-agent-scheduler / main
11:34:17-803172 DEBUG    Running extension installer: E:\AI Software\automatic\extensions-builtin\sd-webui-agent-scheduler\install.py
11:34:18-209526 DEBUG    Submodule: extensions-builtin\sd-webui-controlnet / main
11:34:18-836586 DEBUG    Running extension installer: E:\AI Software\automatic\extensions-builtin\sd-webui-controlnet\install.py
11:34:19-241851 DEBUG    Submodule: extensions-builtin\stable-diffusion-webui-images-browser / main
11:34:19-841680 DEBUG    Running extension installer: E:\AI Software\automatic\extensions-builtin\stable-diffusion-webui-images-browser\install.py
11:34:20-248224 DEBUG    Submodule: extensions-builtin\stable-diffusion-webui-rembg / master
11:34:20-838596 DEBUG    Running extension installer: E:\AI Software\automatic\extensions-builtin\stable-diffusion-webui-rembg\install.py
11:34:21-851891 DEBUG    Extensions all: ['adetailer', 'Auto-Photoshop-StableDiffusion-Plugin', 'openpose-editor', 'sd-civitai-browser-plus', 'sd-ratio-lock', 'sd-webui-animatediff',
                         'sd-webui-openpose-editor', 'sd-webui-prompt-all-in-one', 'sd-webui-prompt-format', 'sd-webui-reactor', 'sd_civitai_extension', 'ultimate-upscale-for-automatic1111']
11:34:21-903170 DEBUG    Submodule: extensions\adetailer / main
11:34:22-481925 DEBUG    Running extension installer: E:\AI Software\automatic\extensions\adetailer\install.py
11:34:22-901767 DEBUG    Submodule: extensions\Auto-Photoshop-StableDiffusion-Plugin / master
11:34:23-516355 DEBUG    Running extension installer: E:\AI Software\automatic\extensions\Auto-Photoshop-StableDiffusion-Plugin\install.py
11:34:25-055969 DEBUG    Submodule: extensions\openpose-editor / master
11:34:25-815902 DEBUG    Submodule: extensions\sd-civitai-browser-plus / main
11:34:26-392001 DEBUG    Running extension installer: E:\AI Software\automatic\extensions\sd-civitai-browser-plus\install.py
11:34:26-794230 DEBUG    Submodule: extensions\sd-ratio-lock / master
11:34:27-522571 DEBUG    Submodule: extensions\sd-webui-animatediff / master
11:34:28-254330 DEBUG    Submodule: extensions\sd-webui-openpose-editor / main
11:34:28-991211 DEBUG    Submodule: extensions\sd-webui-prompt-all-in-one / main
11:34:29-577807 DEBUG    Running extension installer: E:\AI Software\automatic\extensions\sd-webui-prompt-all-in-one\install.py
11:34:35-048990 DEBUG    Submodule: extensions\sd-webui-prompt-format / main
11:34:35-991038 DEBUG    Submodule: extensions\sd-webui-reactor / main
11:34:36-577907 DEBUG    Running extension installer: E:\AI Software\automatic\extensions\sd-webui-reactor\install.py
11:34:37-003072 DEBUG    Submodule: extensions\sd_civitai_extension / main
11:34:37-606663 DEBUG    Running extension installer: E:\AI Software\automatic\extensions\sd_civitai_extension\install.py
11:34:40-194186 DEBUG    Submodule: extensions\ultimate-upscale-for-automatic1111 / master
11:34:40-903002 INFO     Extensions enabled: ['a1111-sd-webui-lycoris', 'clip-interrogator-ext', 'LDSR', 'Lora', 'multidiffusion-upscaler-for-automatic1111', 'ScuNET', 'sd-extension-chainner',
                         'sd-extension-system-info', 'sd-webui-agent-scheduler', 'sd-webui-controlnet', 'stable-diffusion-webui-images-browser', 'stable-diffusion-webui-rembg', 'SwinIR',
                         'adetailer', 'Auto-Photoshop-StableDiffusion-Plugin', 'openpose-editor', 'sd-civitai-browser-plus', 'sd-ratio-lock', 'sd-webui-animatediff',
                         'sd-webui-openpose-editor', 'sd-webui-prompt-all-in-one', 'sd-webui-prompt-format', 'sd-webui-reactor', 'sd_civitai_extension', 'ultimate-upscale-for-automatic1111']
11:34:40-904998 INFO     Verifying packages
11:34:40-906999 INFO     Updating Wiki
11:34:40-958761 DEBUG    Submodule: E:\AI Software\automatic\wiki / master
11:34:41-581961 DEBUG    Setup complete without errors: 1697625282
11:34:41-589957 INFO     Extension preload: {'extensions-builtin': 0.0, 'extensions': 0.0}
11:34:41-591959 DEBUG    Starting module: <module 'webui' from 'E:\\AI Software\\automatic\\webui.py'>
11:34:41-592957 INFO     Command line args: ['--use-directml', '--autolaunch', '--debug', '--upgrade'] autolaunch=True debug=True upgrade=True use_directml=True
11:34:46-417694 DEBUG    Loaded packages: torch=2.0.0+cpu diffusers=0.21.4 gradio=3.43.2
11:34:46-976294 DEBUG    Reading: config.json len=34
11:34:46-977295 DEBUG    Unknown settings: ['ad_save_images_before', 'civitai_link_key', 'civitai_nsfw_previews', 'multiple_tqdm']
11:34:47-036297 INFO     Engine: backend=Backend.ORIGINAL compute=directml mode=no_grad device=privateuseone:0 cross-optimization="Sub-quadratic"
11:34:47-132105 INFO     Device: device=AMD Radeon VII n=2 directml=0.2.0.dev230426
11:34:47-724939 DEBUG    Entering start sequence
11:34:47-727937 DEBUG    Initializing
11:34:47-730942 INFO     Available VAEs: models\VAE items=4
11:34:47-737942 DEBUG    Reading: cache.json len=2
11:34:47-742942 DEBUG    Reading: metadata.json len=172
11:34:47-746449 INFO     Available models: models\Stable-diffusion items=34 time=0.01s
11:34:47-925967 DEBUG    Loading extensions
11:34:50-081765 INFO     Extension: script='extensions-builtin\sd-webui-controlnet\scripts\controlnet.py' Warning: ControlNet failed to load SGM - will use LDM instead.
11:34:50-082765 INFO     Extension: script='extensions-builtin\sd-webui-controlnet\scripts\controlnet.py' ControlNet preprocessor location: E:\AI
                         Software\automatic\extensions-builtin\sd-webui-controlnet\annotator\downloads
11:34:50-090769 INFO     Extension: script='extensions-builtin\sd-webui-controlnet\scripts\hook.py' Warning: ControlNet failed to load SGM - will use LDM instead.
11:34:51-666523 INFO     Extension: script='extensions\adetailer\scripts\!adetailer.py' [-] ADetailer initialized. version: 23.10.1, num models: 9
11:34:51-712524 INFO     Extension: script='extensions\Auto-Photoshop-StableDiffusion-Plugin\scripts\main.py' python_server_full_path:
                         extensions\Auto-Photoshop-StableDiffusion-Plugin\server/python_server
11:34:54-772724 INFO     Extension: script='extensions\sd-civitai-browser-plus\scripts\civitai_api.py' Aria2 RPC started
11:34:56-049742 INFO     Extension: script='extensions\sd-webui-prompt-all-in-one\scripts\on_app_started.py' sd-webui-prompt-all-in-one background API service started successfully.
11:34:56-274027 ERROR    Module load: extensions\sd-webui-reactor\scripts\reactor_api.py: ImportError
╭───────────────────────────────────────────────────────────────────────────── Traceback (most recent call last) ──────────────────────────────────────────────────────────────────────────────╮
│ E:\AI Software\automatic\modules\script_loading.py:21 in load_module                                                                                                                         │
│                                                                                                                                                                                              │
│   20 │   │   │   with contextlib.redirect_stdout(io.StringIO()) as stdout:                                                                                                                   │
│ ❱ 21 │   │   │   │   module_spec.loader.exec_module(module)                                                                                                                                  │
│   22 │   │   │   setup_logging() # reset since scripts can hijaack logging                                                                                                                   │
│ in exec_module:883                                                                                                                                                                           │
│ in _call_with_frames_removed:241                                                                                                                                                             │
│                                                                                                                                                                                              │
│ E:\AI Software\automatic\extensions\sd-webui-reactor\scripts\reactor_api.py:17 in <module>                                                                                                   │
│                                                                                                                                                                                              │
│    16                                                                                                                                                                                        │
│ ❱  17 from scripts.reactor_swapper import EnhancementOptions, swap_face                                                                                                                      │
│    18 from scripts.reactor_logger import logger                                                                                                                                              │
│                                                                                                                                                                                              │
│ E:\AI Software\automatic\extensions\sd-webui-reactor\scripts\reactor_swapper.py:13 in <module>                                                                                               │
│                                                                                                                                                                                              │
│    12 from modules.face_restoration import FaceRestoration                                                                                                                                   │
│ ❱  13 from modules import codeformer_model                                                                                                                                                   │
│    14 from modules.upscaler import UpscalerData                                                                                                                                              │
╰──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
ImportError: cannot import name 'codeformer_model' from 'modules' (unknown location)
11:34:56-304027 ERROR    Module load: extensions\sd-webui-reactor\scripts\reactor_faceswap.py: ImportError
╭───────────────────────────────────────────────────────────────────────────── Traceback (most recent call last) ──────────────────────────────────────────────────────────────────────────────╮
│ E:\AI Software\automatic\modules\script_loading.py:21 in load_module                                                                                                                         │
│                                                                                                                                                                                              │
│   20 │   │   │   with contextlib.redirect_stdout(io.StringIO()) as stdout:                                                                                                                   │
│ ❱ 21 │   │   │   │   module_spec.loader.exec_module(module)                                                                                                                                  │
│   22 │   │   │   setup_logging() # reset since scripts can hijaack logging                                                                                                                   │
│ in exec_module:883                                                                                                                                                                           │
│ in _call_with_frames_removed:241                                                                                                                                                             │
│                                                                                                                                                                                              │
│ E:\AI Software\automatic\extensions\sd-webui-reactor\scripts\reactor_faceswap.py:26 in <module>                                                                                              │
│                                                                                                                                                                                              │
│    25 from scripts.reactor_logger import logger                                                                                                                                              │
│ ❱  26 from scripts.reactor_swapper import EnhancementOptions, swap_face, check_process_halt, r                                                                                               │
│    27 from scripts.reactor_version import version_flag, app_title                                                                                                                            │
│                                                                                                                                                                                              │
│ E:\AI Software\automatic\extensions\sd-webui-reactor\scripts\reactor_swapper.py:13 in <module>                                                                                               │
│                                                                                                                                                                                              │
│    12 from modules.face_restoration import FaceRestoration                                                                                                                                   │
│ ❱  13 from modules import codeformer_model                                                                                                                                                   │
│    14 from modules.upscaler import UpscalerData                                                                                                                                              │
╰──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
ImportError: cannot import name 'codeformer_model' from 'modules' (unknown location)
11:34:56-314030 ERROR    Module load: extensions\sd-webui-reactor\scripts\reactor_swapper.py: ImportError
╭───────────────────────────────────────────────────────────────────────────── Traceback (most recent call last) ──────────────────────────────────────────────────────────────────────────────╮
│ E:\AI Software\automatic\modules\script_loading.py:21 in load_module                                                                                                                         │
│                                                                                                                                                                                              │
│   20 │   │   │   with contextlib.redirect_stdout(io.StringIO()) as stdout:                                                                                                                   │
│ ❱ 21 │   │   │   │   module_spec.loader.exec_module(module)                                                                                                                                  │
│   22 │   │   │   setup_logging() # reset since scripts can hijaack logging                                                                                                                   │
│ in exec_module:883                                                                                                                                                                           │
│ in _call_with_frames_removed:241                                                                                                                                                             │
│                                                                                                                                                                                              │
│ E:\AI Software\automatic\extensions\sd-webui-reactor\scripts\reactor_swapper.py:13 in <module>                                                                                               │
│                                                                                                                                                                                              │
│    12 from modules.face_restoration import FaceRestoration                                                                                                                                   │
│ ❱  13 from modules import codeformer_model                                                                                                                                                   │
│    14 from modules.upscaler import UpscalerData                                                                                                                                              │
╰──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
ImportError: cannot import name 'codeformer_model' from 'modules' (unknown location)
11:34:56-389646 STATUS   Extension: script='extensions\sd_civitai_extension\scripts\api.py' Civitai: API loaded
11:34:56-657964 STATUS   Extensions time: 8.73s { a1111-sd-webui-lycoris=0.89s Lora=0.13s sd-extension-chainner=0.06s sd-webui-agent-scheduler=0.29s sd-webui-controlnet=0.71s
                         stable-diffusion-webui-images-browser=0.07s stable-diffusion-webui-rembg=0.85s adetailer=0.65s sd-civitai-browser-plus=4.09s sd-ratio-lock=0.05s
                         sd-webui-animatediff=0.09s sd-webui-prompt-all-in-one=0.05s sd-webui-reactor=0.24s sd_civitai_extension=0.34s }
11:34:56-730961 DEBUG    Reading: html/upscalers.json len=4
11:34:56-733961 DEBUG    Loaded upscalers: total=41 downloaded=1 user=0 ['None', 'Lanczos', 'Nearest', 'ChaiNNer', 'ESRGAN', 'LDSR', 'RealESRGAN', 'SCUNet', 'SD', 'SwinIR']
11:34:56-819446 STATUS   Loading UI theme: name=black-teal style=Auto
11:34:56-824451 DEBUG    Loaded styles: folder=models\styles items=288
11:34:56-828451 DEBUG    Creating UI
11:34:56-974638 DEBUG    Extra networks: page='model' items=34 subdirs=2 tab=txt2img dirs=['models\\Stable-diffusion', 'models\\Diffusers', 'E:\\AI
                         Software\\automatic\\models\\Stable-diffusion'] time=0.05s
11:34:56-986638 DEBUG    Extra networks: page='style' items=288 subdirs=2 tab=txt2img dirs=['models\\styles', 'html'] time=0.0s
11:34:56-989640 DEBUG    Extra networks: page='embedding' items=12 subdirs=1 tab=txt2img dirs=['models\\embeddings'] time=0.03s
11:34:56-991642 DEBUG    Extra networks: page='hypernetwork' items=0 subdirs=0 tab=txt2img dirs=['models\\hypernetworks'] time=0.0s
11:34:56-993643 DEBUG    Extra networks: page='vae' items=4 subdirs=1 tab=txt2img dirs=['models\\VAE'] time=0.01s
11:34:57-003642 DEBUG    Extra networks: page='lora' items=138 subdirs=1 tab=txt2img dirs=['models\\Lora', 'models\\LyCORIS'] time=0.11s
11:34:57-005641 DEBUG    Extra networks: page='lycoris' items=0 subdirs=0 tab=txt2img dirs=['models\\LyCORIS'] time=0.0s
11:34:58-405126 DEBUG    Reading: ui-config.json len=3
11:34:58-467306 DEBUG    Themes: builtin=7 default=5 external=55
11:34:59-053511 DEBUG    Script: 0.3s ui_tabs E:\AI Software\automatic\extensions-builtin\stable-diffusion-webui-images-browser\scripts\image_browser.py
11:34:59-106512 DEBUG    Reading: E:\AI Software\automatic\html\extensions.json len=241
11:35:00-614149 DEBUG    Extension list: processed=233 installed=25 enabled=25 disabled=0 visible=233 hidden=0
11:35:03-026719 STATUS   Local URL: http://127.0.0.1:7860/
11:35:03-029719 DEBUG    Gradio registered functions: 2349
11:35:03-030718 STATUS   Initializing middleware
11:35:03-033718 DEBUG    Creating API
11:35:03-205953 STATUS   [AgentScheduler] Task queue is empty
11:35:03-206953 STATUS   [AgentScheduler] Registering APIs
11:35:04-020099 DEBUG    Script: 0.8s app_started E:\AI Software\automatic\extensions\sd-webui-prompt-all-in-one\scripts\on_app_started.py
Civitai: Check resources for missing info files
Civitai Link: Connecting to Civitai Link Server
Civitai: Found 0 resources missing info files
Civitai: No info found on Civitai
Civitai Link: Connected to Civitai Link Server
11:35:05-531894 DEBUG    Script: 1.51s app_started E:\AI Software\automatic\extensions\sd_civitai_extension\scripts\link.py
Civitai: Check resources for missing preview images
11:35:05-648040 DEBUG    Scripts setup: ['X/Y/Z Grid:0.006s', 'Tiled Diffusion:0.029s', 'ControlNet:0.06s', 'ADetailer:1.155s', 'AnimateDiff:0.006s']
11:35:05-650220 DEBUG    Model metadata: metadata.json no changes
11:35:05-787844 STATUS   Select: model="revAnimated_v122EOL [4199bcdd14]"
11:35:05-789845 DEBUG    Load model weights: existing=False target=E:\AI Software\automatic\models\Stable-diffusion\revAnimated_v122EOL.safetensors info=None
Loading weights: E:\AI Software\automatic\models\Stable-diffusion\revAnimated_v122EOL.safetensors ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/5.5 GB -:--:--
11:35:05-874361 DEBUG    Load model: name=E:\AI Software\automatic\models\Stable-diffusion\revAnimated_v122EOL.safetensors dict=True
11:35:05-881359 DEBUG    Desired Torch parameters: dtype=FP16 no-half=False no-half-vae=False upscast=False
11:35:05-882360 STATUS   Setting Torch parameters: device=privateuseone:0 dtype=torch.float16 vae=torch.float16 unet=torch.float16 context=no_grad fp16=True bf16=False
11:35:05-883361 DEBUG    Model dict loaded: {'ram': {'used': 0.91, 'total': 63.91}, 'gpu': {'used': 0.01, 'total': 0.01}, 'retries': 0, 'oom': 0}
11:35:05-894358 DEBUG    Model config loaded: {'ram': {'used': 0.91, 'total': 63.91}, 'gpu': {'used': 0.01, 'total': 0.01}, 'retries': 0, 'oom': 0}
11:35:06-257216 STATUS   LDM: LatentDiffusion: Running in eps-prediction mode
11:35:06-259214 STATUS   LDM: Civitai Link: Joined room ddb9823c942c0dfcb9405bcf8f8ced0477f3b871aa311b1239519869a1b021c8069d2afbc65bcf355518e778a3e4877da1e822ca32d4270e1cbd440307b04cc5
11:35:06-260215 STATUS   LDM: Civitai Link: Presence update: SD: 1, Clients: 1
11:35:06-261215 STATUS   LDM: Civitai Link: Connected to Civitai Instance
11:35:06-262214 STATUS   LDM: Civitai Link: Incoming Command: activities:list
11:35:06-263213 STATUS   LDM: DiffusionWrapper has 859.52 M params.
11:35:06-264215 STATUS   LDM: Civitai: Found 0 resources missing preview images
11:35:06-265217 STATUS   LDM: Civitai: No preview images found on Civitai
11:35:06-265217 STATUS   LDM: Civitai Link: Incoming Command: resources:list
11:35:06-266213 DEBUG    Model created from config: E:\AI Software\automatic\configs\v1-inference.yaml
11:35:06-267214 STATUS   Autodetect: model="Stable Diffusion" class=StableDiffusionPipeline file="E:\AI Software\automatic\models\Stable-diffusion\revAnimated_v122EOL.safetensors" size=5250MB
11:35:06-268214 DEBUG    Model weights loading: {'ram': {'used': 1.87, 'total': 63.91}, 'gpu': {'used': 0.01, 'total': 0.01}, 'retries': 0, 'oom': 0}
11:35:07-354043 DEBUG    Model weights loaded: {'ram': {'used': 8.05, 'total': 63.91}, 'gpu': {'used': 0.01, 'total': 0.01}, 'retries': 0, 'oom': 0}
11:35:08-669613 DEBUG    Model weights moved: {'ram': {'used': 6.31, 'total': 63.91}, 'gpu': {'used': 2.08, 'total': 2.34}, 'retries': 0, 'oom': 0}
11:35:08-676122 STATUS   Cross-attention: optimization=Sub-quadratic options=[]
11:35:08-697124 STATUS   Loaded embeddings: loaded=12 skipped=0 time=0.01s
11:35:08-705984 STATUS   Model loaded in 2.92s { load=0.09s create=0.37s apply=0.62s vae=0.46s move=1.31s }
11:35:09-064267 DEBUG    gc: collected=5174 device=privateuseone:0 {'ram': {'used': 6.32, 'total': 63.91}, 'gpu': {'used': 2.08, 'total': 2.34}, 'retries': 0, 'oom': 0}
11:35:09-066267 STATUS   Model load finished: {'ram': {'used': 6.32, 'total': 63.91}, 'gpu': {'used': 2.08, 'total': 2.34}, 'retries': 0, 'oom': 0} cached=0
11:35:09-515510 DEBUG    Saving: config.json len=1424
11:35:09-517516 DEBUG    Unused settings: ['multiple_tqdm']
11:35:09-519521 STATUS   Startup time: 27.92s { torch=4.19s gradio=0.60s libraries=1.31s extensions=8.73s face-restore=0.18s upscalers=0.08s onchange=0.09s ui-extra-networks=0.18s
                         ui-txt2img=1.22s ui-img2img=0.13s ui-settings=0.34s ui-extensions=1.91s ui-defaults=0.06s launch=2.33s api=0.07s app-started=2.55s checkpoint=3.87s }
11:35:09-523515 STATUS   Launching browser
11:35:12-316915 STATUS   MOTD:
11:35:15-292923 DEBUG    Themes: builtin=7 default=5 external=55
11:35:15-519966 STATUS   Browser session: client=127.0.0.1 agent=Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36
11:35:59-984147 DEBUG    Server alive=True jobs=1 requests=107 uptime=73s memory used=1.2 total=63.91 job="load model" 0/-1
11:37:59-877297 DEBUG    Server alive=True jobs=1 requests=143 uptime=193s memory used=1.2 total=63.91 job="load model" 0/-1
Civitai Link: Disconnected from Civitai Link Server
Civitai Link: Connected to Civitai Link Server
Civitai Link: Joined room ddb9823c942c0dfcb9405bcf8f8ced0477f3b871aa311b1239519869a1b021c8069d2afbc65bcf355518e778a3e4877da1e822ca32d4270e1cbd440307b04cc5
Civitai Link: Presence update: SD: 1, Clients: 1
Civitai Link: Incoming Command: activities:list
Civitai Link: Incoming Command: resources:list
11:39:59-806963 DEBUG    Server alive=True jobs=1 requests=179 uptime=313s memory used=1.2 total=63.91 job="load model" 0/-1
11:41:59-764222 DEBUG    Server alive=True jobs=1 requests=215 uptime=433s memory used=1.19 total=63.91 job="load model" 0/-1
11:43:59-708523 DEBUG    Server alive=True jobs=1 requests=251 uptime=553s memory used=1.19 total=63.91 job="load model" 0/-1
11:45:59-633483 DEBUG    Server alive=True jobs=1 requests=287 uptime=673s memory used=1.19 total=63.91 job="load model" 0/-1
11:47:59-572341 DEBUG    Server alive=True jobs=1 requests=323 uptime=793s memory used=1.19 total=63.91 job="load model" 0/-1
Loading weights: models\VAE\clearvae_v22.safetensors ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/334.6 MB -:--:--
11:49:36-843558 STATUS   Cross-attention: optimization=Sub-quadratic options=[]
11:49:36-851765 STATUS   VAE weights loaded: models\VAE\clearvae_v22.safetensors
11:49:37-127448 DEBUG    Saving: config.json len=1444
11:49:37-128451 DEBUG    Unused settings: ['multiple_tqdm']
11:49:37-129448 DEBUG    Setting changed: key=sd_vae, value=clearvae_v22.safetensors
11:49:59-588446 DEBUG    Server alive=True jobs=1 requests=376 uptime=913s memory used=1.44 total=63.91 job="load model" 0/-1
11:50:32-671357 DEBUG    txt2img: id_task=task(u29wsu1qt4nw7mj)|prompt=(world of warcraft),1girl,solo,elf,(long eyebrows),red eyes,narrow_waist,sitting,(crossed
                         leg),(halloween:1.3),witch_hat,full moon,pumpkin,(bats),<lora:sylvanas-v1:0.6>,(full
                         body),wide_shot,<lora:add_detail:0.5>,|negative_prompt=FastNegativeV2,badhandv4,multiple_girls,multiple_moons,|prompt_styles=[]|steps=20|sampler_index=11|latent_index=
                         None|full_quality=True|restore_faces=False|tiling=False|n_iter=1|batch_size=1|cfg_scale=7|clip_skip=2|seed=-1.0|subseed=-1.0|subseed_strength=0|seed_resize_from_h=0|se
                         ed_resize_from_w=0||height=576|width=1024|enable_hr=False|denoising_strength=0.5|hr_scale=2|hr_upscaler=None|hr_force=False|hr_second_pass_steps=20|hr_resize_x=0|hr_re
                         size_y=0|image_cfg_scale=6|diffusers_guidance_rescale=0.7|refiner_steps=5|refiner_start=0.8|refiner_prompt=|refiner_negative=|override_settings_texts=[]
11:50:32-675356 STATUS   Applying free-u: b1=1.2 b2=1.4 s1=0.9 s2=0.2
11:50:32-677361 WARNING  Hypertile UNet is not compatible with Sub-quadratic cross-attention optimization
Loading weights: models\Lora\sylvanas-v1.safetensors ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/302.1 MB -:--:--
Loading weights: models\Lora\add_detail.safetensors ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/37.9 MB -:--:--
11:50:33-920805 STATUS   Applying LoRA: ['sylvanas-v1', 'add_detail'] patch=0.00s load=1.24s
11:50:34-410125 DEBUG    Sampler: sampler=DPM++ 2M config={'scheduler': 'karras', 'brownian_noise': False}
 10%|███████████████▌                                                                                                                                           | 2/20 [00:10<01:33,  5.18s/it]11:50:45-148420 DEBUG    Loaded VAE decode approximate: model="models\VAE-approx\model.pt"
100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [00:46<00:00,  2.33s/it]

0: 384x640 1 face, 59.1ms
Speed: 2.0ms preprocess, 59.1ms inference, 1.0ms postprocess per image at shape (1, 3, 384, 640)
11:51:23-725623 STATUS   Applying free-u: b1=1.2 b2=1.4 s1=0.9 s2=0.2
11:51:23-726622 WARNING  Hypertile UNet is not compatible with Sub-quadratic cross-attention optimization
11:51:23-728622 DEBUG    Sampler: sampler=DPM++ 2M config={'scheduler': 'karras', 'brownian_noise': False}
Loading weights: models\Lora\sylvanas-v1.safetensors ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/302.1 MB -:--:--
Loading weights: models\Lora\add_detail.safetensors ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/37.9 MB -:--:--
11:51:25-160179 STATUS   Applying LoRA: ['sylvanas-v1', 'add_detail'] patch=0.00s load=1.35s
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 9/9 [00:17<00:00,  1.92s/it]
11:51:43-702747 STATUS   Processed: images=1 time=19.98s its=1.00 memory={'ram': {'used': 4.13, 'total': 63.91}, 'gpu': {'used': 14.58, 'total': 15.29}, 'retries': 0, 'oom': 0}
11:51:43-708752 DEBUG    Saving: image="outputs\text\00161-revAnimated_v122EOL-world of warcraft 1girl solo elf long eyebrows-ad-before.webp" type=WEBP size=1024x576
11:51:44-620990 DEBUG    Saving: image="outputs\text\00162-revAnimated_v122EOL-world of warcraft 1girl solo elf long eyebrows.webp" type=WEBP size=1024x576
11:51:45-349358 STATUS   Processed: images=1 time=72.67s its=0.28 memory={'ram': {'used': 4.11, 'total': 63.91}, 'gpu': {'used': 14.58, 'total': 15.29}, 'retries': 0, 'oom': 0}
11:52:00-471974 DEBUG    Server alive=True jobs=1 requests=545 uptime=1034s memory used=4.1 total=63.91 job="load model" 0/0
11:54:00-290426 DEBUG    Server alive=True jobs=1 requests=581 uptime=1153s memory used=4.1 total=63.91 job="load model" 0/0

List of installed extensions

No response

Bing-su commented 11 months ago

I think this is an issue with direct ml pytorch.

Fot3k commented 11 months ago

I think this is an issue with direct ml pytorch.

Finally fixed it, needs full precision enabled. image