search

vladmandic / automatic

SD.Next: Advanced Implementation of Stable Diffusion and other Diffusion-based generative image models
https://github.com/vladmandic/automatic
GNU Affero General Public License v3.0
5.61k stars 412 forks source link

[Issue]: Lora persisting after removal #3491

Closed zaxwashere closed 3 days ago

zaxwashere commented 4 days ago

Issue Description

Style loras for ponyxl models seem to persist until a model reload is applied. I've attached a log on a fresh venv with most addons disabled. I have also attached 3 pngs to show the issue visually.

This occurs at any lora strength. For ease of confirmation, I set a minimalist artstyle lora at 1.5 strength. This still occurs at 0.4.

Steps: Load model input base prompt, generate add lora to prompt, generate 2nd image delete lora from prompt, generate 3rd image.

The 3rd image will still show the lora's effects. I've tried disabling prompt cache, disabling timesteps, different models, different loras. The issue persists until i change models over.

attached are 3 images, and my config.json in case that's helpful. Thanks! copied_config.json

No lora, 1st generation no lora 1st gen Lora on, 2nd generation lora on 2nd gen Lora removed, 3rd generation lora off 3rd gen

Version Platform Description

Windows 11, 23H2 amd 5700x3d 64 gb drr4, 3600mhz rtx 3060 12 gb, driver 565.90 python 3.10.6 SDNEXT dev branch, db3e03cb System packages: {'torch': '2.4.1+cu124', 'diffusers': '0.31.0.dev0', 'gradio': '3.43.2', 'transformers': '4.45.2', 'accelerate': '1.0.0'}

Relevant log output

2024-10-16 14:22:10,617 | sd | INFO | launch | Starting SD.Next
2024-10-16 14:22:10,620 | sd | INFO | installer | Logger: file="C:\Users\zaxof\OneDrive\Documents\GitHub\nvidia_sdnext\sdnext.log" level=DEBUG size=65 mode=create
2024-10-16 14:22:10,621 | sd | INFO | installer | Python: version=3.10.6 platform=Windows bin="C:\Users\zaxof\OneDrive\Documents\GitHub\nvidia_sdnext\venv\Scripts\python.exe" venv="C:\Users\zaxof\OneDrive\Documents\GitHub\nvidia_sdnext\venv"
2024-10-16 14:22:10,805 | sd | INFO | installer | Version: app=sd.next updated=2024-10-16 hash=db3e03cb branch=dev url=https://github.com/vladmandic/automatic.git/tree/dev ui=dev
2024-10-16 14:22:11,140 | sd | INFO | installer | Updating main repository
2024-10-16 14:22:11,933 | sd | INFO | installer | Repository upgraded: db3e03cb Wed Oct 16 12:10:43 2024 -0400
2024-10-16 14:22:11,940 | sd | INFO | launch | Platform: arch=AMD64 cpu=AMD64 Family 25 Model 33 Stepping 2, AuthenticAMD system=Windows release=Windows-10-10.0.22631-SP0 python=3.10.6
2024-10-16 14:22:11,942 | sd | DEBUG | installer | Setting environment tuning
2024-10-16 14:22:11,943 | sd | DEBUG | installer | Torch allocator: "garbage_collection_threshold:0.90,max_split_size_mb:512"
2024-10-16 14:22:11,953 | sd | DEBUG | installer | Torch overrides: cuda=True rocm=False ipex=False diml=False openvino=False zluda=False
2024-10-16 14:22:11,955 | sd | INFO | installer | CUDA: nVidia toolkit detected
2024-10-16 14:22:12,121 | sd | INFO | installer | Verifying requirements
2024-10-16 14:22:12,125 | sd | INFO | installer | Verifying packages
2024-10-16 14:22:12,126 | sd | INFO | launch | Startup: standard
2024-10-16 14:22:12,127 | sd | INFO | installer | Verifying submodules
2024-10-16 14:22:14,067 | sd | DEBUG | installer | Git submodule: extensions-builtin/sd-extension-chainner / main
2024-10-16 14:22:14,614 | sd | DEBUG | installer | Git submodule: extensions-builtin/sd-extension-system-info / main
2024-10-16 14:22:15,147 | sd | DEBUG | installer | Git submodule: extensions-builtin/sd-webui-agent-scheduler / main
2024-10-16 14:22:15,688 | sd | DEBUG | installer | Git submodule: extensions-builtin/sdnext-modernui / dev
2024-10-16 14:22:16,228 | sd | DEBUG | installer | Git submodule: extensions-builtin/stable-diffusion-webui-rembg / master
2024-10-16 14:22:16,785 | sd | DEBUG | installer | Git submodule: modules/k-diffusion / master
2024-10-16 14:22:17,321 | sd | DEBUG | installer | Git submodule: wiki / master
2024-10-16 14:22:17,893 | sd | DEBUG | paths | Register paths
2024-10-16 14:22:17,987 | sd | DEBUG | installer | Installed packages: 218
2024-10-16 14:22:17,988 | sd | DEBUG | installer | Extensions all: ['Lora', 'sd-extension-chainner', 'sd-extension-system-info', 'stable-diffusion-webui-rembg']
2024-10-16 14:22:18,530 | sd | DEBUG | installer | Git submodule: extensions-builtin\sd-extension-chainner / main
2024-10-16 14:22:19,147 | sd | DEBUG | installer | Git submodule: extensions-builtin\sd-extension-system-info / main
2024-10-16 14:22:19,765 | sd | DEBUG | installer | Git submodule: extensions-builtin\stable-diffusion-webui-rembg / master
2024-10-16 14:22:20,327 | sd | DEBUG | installer | Extension installer: C:\Users\zaxof\OneDrive\Documents\GitHub\nvidia_sdnext\extensions-builtin\stable-diffusion-webui-rembg\install.py
2024-10-16 14:22:27,105 | sd | DEBUG | installer | Extensions all: ['a1111-sd-webui-tagcomplete', 'adetailer', 'sd-civitai-browser-plus_fix', 'sd-webui-infinite-image-browsing']
2024-10-16 14:22:27,139 | sd | DEBUG | installer | Git submodule: extensions\a1111-sd-webui-tagcomplete / main
2024-10-16 14:22:27,794 | sd | DEBUG | installer | Git detached head detected: folder="extensions\adetailer" reattach=main
2024-10-16 14:22:27,795 | sd | DEBUG | installer | Git submodule: extensions\adetailer / main
2024-10-16 14:22:28,381 | sd | DEBUG | installer | Extension force: name="adetailer" commit=a89c01d
2024-10-16 14:22:28,382 | sd | DEBUG | installer | Extension installer: C:\Users\zaxof\OneDrive\Documents\GitHub\nvidia_sdnext\extensions\adetailer\install.py
2024-10-16 14:22:28,647 | sd | DEBUG | installer | Git submodule: extensions\sd-civitai-browser-plus_fix / main
2024-10-16 14:22:29,170 | sd | DEBUG | installer | Extension installer: C:\Users\zaxof\OneDrive\Documents\GitHub\nvidia_sdnext\extensions\sd-civitai-browser-plus_fix\install.py
2024-10-16 14:22:39,871 | sd | DEBUG | installer | Git submodule: extensions\sd-webui-infinite-image-browsing / main
2024-10-16 14:22:40,396 | sd | DEBUG | installer | Extension installer: C:\Users\zaxof\OneDrive\Documents\GitHub\nvidia_sdnext\extensions\sd-webui-infinite-image-browsing\install.py
2024-10-16 14:22:59,559 | sd | INFO | installer | Extensions enabled: ['Lora', 'sd-extension-chainner', 'sd-extension-system-info', 'stable-diffusion-webui-rembg', 'a1111-sd-webui-tagcomplete', 'adetailer', 'sd-civitai-browser-plus_fix', 'sd-webui-infinite-image-browsing']
2024-10-16 14:22:59,560 | sd | INFO | installer | Verifying requirements
2024-10-16 14:22:59,562 | sd | INFO | installer | Updating Wiki
2024-10-16 14:22:59,596 | sd | DEBUG | installer | Git submodule: C:\Users\zaxof\OneDrive\Documents\GitHub\nvidia_sdnext\wiki / master
2024-10-16 14:23:00,112 | sd | DEBUG | launch | Setup complete without errors: 1729102980
2024-10-16 14:23:00,117 | sd | DEBUG | installer | Extension preload: {'extensions-builtin': 0.0, 'extensions': 0.0}
2024-10-16 14:23:00,118 | sd | DEBUG | launch | Starting module: <module 'webui' from 'C:\\Users\\zaxof\\OneDrive\\Documents\\GitHub\\nvidia_sdnext\\webui.py'>
2024-10-16 14:23:00,120 | sd | INFO | launch | Command line args: ['--debug', '--insecure', '--listen', '--use-cuda', '--debug', '--upgrade'] insecure=True listen=True upgrade=True use_cuda=True debug=True
2024-10-16 14:23:00,121 | sd | DEBUG | launch | Env flags: []
2024-10-16 14:23:05,999 | sd | INFO | loader | System packages: {'torch': '2.4.1+cu124', 'diffusers': '0.31.0.dev0', 'gradio': '3.43.2', 'transformers': '4.45.2', 'accelerate': '1.0.0'}
2024-10-16 14:23:06,539 | sd | DEBUG | shared | Huggingface cache: folder="C:\Users\zaxof\.cache\huggingface\hub"
2024-10-16 14:23:06,633 | sd | INFO | shared | Device detect: memory=12.0 optimization=none
2024-10-16 14:23:06,636 | sd | DEBUG | shared | Read: file="config.json" json=64 bytes=2812 time=0.000
2024-10-16 14:23:06,637 | sd | DEBUG | shared | Unknown settings: ['disable_dns', 'show_log', 'unpack_zip', 'custom_api_key', 'use_LORA', 'multiple_tqdm', 'ad_save_previews', 'ad_save_images_before', 'facehires_strength']
2024-10-16 14:23:06,639 | sd | INFO | shared | Engine: backend=Backend.DIFFUSERS compute=cuda device=cuda attention="Scaled-Dot-Product" mode=no_grad
2024-10-16 14:23:06,640 | sd | DEBUG | shared | Read: file="html\reference.json" json=55 bytes=30515 time=0.001
2024-10-16 14:23:06,679 | sd | INFO | devices | Torch parameters: backend=cuda device=cuda config=Auto dtype=torch.bfloat16 vae=torch.bfloat16 unet=torch.bfloat16 context=no_grad nohalf=False nohalfvae=False upscast=False deterministic=False test-fp16=True test-bf16=True optimization="Scaled-Dot-Product"
2024-10-16 14:23:07,042 | sd | DEBUG | __init__ | ONNX: version=1.19.2 provider=CUDAExecutionProvider, available=['AzureExecutionProvider', 'CPUExecutionProvider']
2024-10-16 14:23:07,162 | sd | INFO | shared | Device: device=NVIDIA GeForce RTX 3060 n=1 arch=sm_90 capability=(8, 6) cuda=12.4 cudnn=90100 driver=565.90
2024-10-16 14:23:07,259 | sd | DEBUG | sd_hijack | Importing LDM
2024-10-16 14:23:07,274 | sd | DEBUG | webui | Entering start sequence
2024-10-16 14:23:07,278 | sd | DEBUG | webui | Initializing
2024-10-16 14:23:07,318 | sd | INFO | sd_vae | Available VAEs: path="C:\Users\zaxof\OneDrive\Documents\GitHub\nvidia_sdnext\models\VAE" items=4
2024-10-16 14:23:07,320 | sd | INFO | sd_unet | Available UNets: path="F:\vlado\models\UNET" items=1
2024-10-16 14:23:07,321 | sd | INFO | model_te | Available TEs: path="F:\vlado\models\Text-encoder" items=7
2024-10-16 14:23:07,323 | sd | INFO | extensions | Disabled extensions: ['sd-webui-agent-scheduler', 'sdnext-modernui']
2024-10-16 14:23:07,326 | sd | DEBUG | shared | Read: file="cache.json" json=2 bytes=9003 time=0.000
2024-10-16 14:23:07,351 | sd | DEBUG | shared | Read: file="metadata.json" json=1802 bytes=7888407 time=0.024
2024-10-16 14:23:07,361 | sd | DEBUG | modelloader | Scanning diffusers cache: folder="F:\vlado\models\Diffusers" items=4 time=0.00
2024-10-16 14:23:07,363 | sd | INFO | sd_models | Available Models: path="F:\vlado\models\Stable-diffusion" items=32 time=0.04
2024-10-16 14:23:07,436 | sd | INFO | yolo | Available Yolo: path="F:\vlado\models\yolo items=6 downloaded=1
2024-10-16 14:23:07,437 | sd | DEBUG | webui | Load extensions
2024-10-16 14:23:07,646 | sd | INFO | networks | Available LoRAs: items=1717 folders=3
2024-10-16 14:23:07,651 | sd | INFO | script_loading | Extension: script='extensions-builtin\Lora\scripts\lora_script.py' 14:23:07-646600 INFO     Available LoRAs: items=1717 folders=3
2024-10-16 14:23:08,276 | sd | INFO | script_loading | Extension: script='extensions\a1111-sd-webui-tagcomplete\scripts\tag_autocomplete_helper.py' Tag Autocomplete: Could not locate model-keyword extension, Lora trigger word completion will be limited to those added through the extra networks menu.
2024-10-16 14:23:08,518 | sd | INFO | script_loading | Extension: script='extensions\adetailer\scripts\!adetailer.py' [-] ADetailer initialized. version: 24.5.1, num models: 17
2024-10-16 14:23:08,537 | sd | INFO | script_loading | Extension: script='extensions\sd-civitai-browser-plus_fix\scripts\civitai_api.py' CivitAI Browser+: Aria2 RPC started
2024-10-16 14:23:08,596 | sd | DEBUG | webui | Extensions init time: 1.16 Lora=0.15 stable-diffusion-webui-rembg=0.29 a1111-sd-webui-tagcomplete=0.29 adetailer=0.24 sd-civitai-browser-plus_fix=0.05
2024-10-16 14:23:08,611 | sd | DEBUG | shared | Read: file="html/upscalers.json" json=4 bytes=2672 time=0.000
2024-10-16 14:23:08,613 | sd | DEBUG | shared | Read: file="extensions-builtin\sd-extension-chainner\models.json" json=24 bytes=2719 time=0.000
2024-10-16 14:23:08,616 | sd | DEBUG | chainner_model | chaiNNer models: path="F:\vlado\models\chaiNNer" defined=24 discovered=0 downloaded=5
2024-10-16 14:23:08,622 | sd | INFO | modelloader | Available Upscalers: items=53 downloaded=9 user=0 time=0.02 types=['None', 'Lanczos', 'Nearest', 'ChaiNNer', 'AuraSR', 'ESRGAN', 'LDSR', 'RealESRGAN', 'SCUNet', 'SD', 'SwinIR']
2024-10-16 14:23:08,626 | sd | INFO | styles | Available Styles: folder="models\styles" items=3 time=0.00
2024-10-16 14:23:08,630 | sd | DEBUG | webui | Creating UI
2024-10-16 14:23:08,631 | sd | DEBUG | theme | UI themes available: type=Standard themes=12
2024-10-16 14:23:08,632 | sd | INFO | theme | UI theme: type=Standard name="simple-dark"
2024-10-16 14:23:08,643 | sd | DEBUG | ui_javascript | UI theme: css="C:\Users\zaxof\OneDrive\Documents\GitHub\nvidia_sdnext\javascript\simple-dark.css" base="sdnext.css" user="None"
2024-10-16 14:23:08,646 | sd | DEBUG | ui_txt2img | UI initialize: txt2img
2024-10-16 14:23:09,560 | sd | DEBUG | ui_extra_networks | Networks: page='model' items=86 subfolders=2 tab=txt2img folders=['F:\\vlado\\models\\Stable-diffusion', 'F:\\vlado\\models\\Diffusers', 'models\\Reference'] list=0.13 thumb=0.06 desc=0.01 info=0.14 workers=8 sort=Date [Newest]
2024-10-16 14:23:09,600 | sd | DEBUG | ui_extra_networks | Networks: page='lora' items=1717 subfolders=14 tab=txt2img folders=['F:\\vlado\\models\\Lora', 'F:\\vlado\\models\\LyCORIS'] list=0.90 thumb=0.31 desc=0.55 info=2.14 workers=8 sort=Date [Newest]
2024-10-16 14:23:09,606 | sd | DEBUG | ui_extra_networks | Networks: page='style' items=3 subfolders=0 tab=txt2img folders=['models\\styles', 'html'] list=0.00 thumb=0.00 desc=0.00 info=0.00 workers=8 sort=Date [Newest]
2024-10-16 14:23:09,609 | sd | DEBUG | ui_extra_networks | Networks: page='embedding' items=14 subfolders=0 tab=txt2img folders=['F:\\vlado\\models\\embeddings'] list=0.07 thumb=0.01 desc=0.00 info=0.04 workers=8 sort=Date [Newest]
2024-10-16 14:23:09,611 | sd | DEBUG | ui_extra_networks | Networks: page='vae' items=4 subfolders=0 tab=txt2img folders=['C:\\Users\\zaxof\\OneDrive\\Documents\\GitHub\\nvidia_sdnext\\models\\VAE'] list=0.04 thumb=0.03 desc=0.00 info=0.00 workers=8 sort=Date [Newest]
2024-10-16 14:23:09,613 | sd | DEBUG | ui_extra_networks | Networks: page='history' items=0 subfolders=0 tab=txt2img folders=[] list=0.00 thumb=0.00 desc=0.00 info=0.00 workers=8 sort=Date [Newest]
2024-10-16 14:23:09,763 | sd | DEBUG | ui_img2img | UI initialize: img2img
2024-10-16 14:23:09,933 | sd | DEBUG | ui_control_helpers | UI initialize: control models=F:\vlado\models\control
2024-10-16 14:23:10,643 | sd | DEBUG | shared | Read: file="ui-config.json" json=12 bytes=711 time=0.000
2024-10-16 14:23:10,793 | sd | DEBUG | theme | UI themes available: type=Standard themes=12
2024-10-16 14:23:11,875 | sd | DEBUG | ui_extensions | Extension list: processed=391 installed=10 enabled=8 disabled=2 visible=391 hidden=0
2024-10-16 14:23:12,029 | sd | DEBUG | webui | Root paths: ['C:\\Users\\zaxof\\OneDrive\\Documents\\GitHub\\nvidia_sdnext']
2024-10-16 14:23:16,517 | sd | INFO | webui | Local URL: http://localhost:7860/
2024-10-16 14:23:16,518 | sd | DEBUG | webui | Gradio functions: registered=2552
2024-10-16 14:23:16,520 | sd | DEBUG | middleware | FastAPI middleware: ['Middleware', 'Middleware']
2024-10-16 14:23:16,524 | sd | DEBUG | webui | Creating API
2024-10-16 14:23:16,719 | sd | DEBUG | webui | Scripts setup: ['IP Adapters:0.03', 'XYZ Grid:0.031', 'ADetailer:0.091', 'Face:0.018', 'AnimateDiff:0.008', 'CogVideoX:0.01', 'Ctrl-X:0.007', 'LUT Color grading:0.007', 'Image-to-Video:0.009', 'Stable Video Diffusion:0.007']
2024-10-16 14:23:16,721 | sd | DEBUG | sd_models | Model metadata: file="metadata.json" no changes
2024-10-16 14:23:16,722 | sd | DEBUG | webui | Model auto load disabled
2024-10-16 14:23:16,723 | sd | DEBUG | script_callbacks | Script callback init time: system-info.py:app_started=0.07
2024-10-16 14:23:16,725 | sd | INFO | webui | Startup time: 16.60 torch=4.79 gradio=0.82 diffusers=0.10 libraries=1.42 extensions=1.16 detailer=0.07 ui-networks=1.11 ui-txt2img=0.13 ui-img2img=0.13 ui-control=0.20 ui-models=0.21 ui-gallery=0.18 ui-settings=0.38 ui-extensions=0.90 ui-defaults=0.07 launch=4.54 api=0.10 app-started=0.10
2024-10-16 14:23:16,727 | sd | DEBUG | shared | Save: file="config.json" json=64 bytes=2725 time=0.004
2024-10-16 14:23:16,730 | sd | DEBUG | shared | Unused settings: ['multiple_tqdm', 'facehires_strength']
2024-10-16 14:24:00,081 | sd | DEBUG | launch | Server: alive=True jobs=0 requests=13 uptime=54 memory=1.03/63.9 backend=Backend.DIFFUSERS state=idle
2024-10-16 14:24:01,848 | sd | ERROR | middleware | API error: GET: http://73.148.235.221:7860/file=placeholder {'error': 'HTTPException', 'code': 404, 'detail': 'File not found: placeholder.', 'body': '', 'errors': '404: File not found: placeholder.'}
2024-10-16 14:24:02,857 | sd | DEBUG | theme | UI themes available: type=Standard themes=12
2024-10-16 14:24:07,467 | sd | INFO | api | Browser session: user=None client=77.111.246.41 agent=Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.0.0 Safari/537.36 OPR/113.0.0.0
2024-10-16 14:25:19,466 | sd | DEBUG | ui_sections | Sampler set options: timesteps=
2024-10-16 14:25:19,469 | sd | DEBUG | shared | Unused settings: ['multiple_tqdm', 'facehires_strength']
2024-10-16 14:25:22,201 | sd | DEBUG | ui_sections | Sampler set options: sigma=karras
2024-10-16 14:25:22,204 | sd | DEBUG | shared | Unused settings: ['multiple_tqdm', 'facehires_strength']
2024-10-16 14:25:35,889 | sd | DEBUG | modeldata | Model requested: fn=f:txt2img
2024-10-16 14:25:35,891 | sd | INFO | sd_models | Load model: select="novaAnimeXL_ponyV40 [e87c759138]"
2024-10-16 14:25:35,892 | sd | DEBUG | sd_models | Load model: target="F:\vlado\models\Stable-diffusion\novaAnimeXL_ponyV40.safetensors" existing=False info=None
2024-10-16 14:25:35,894 | sd | DEBUG | sd_models | Load model: path="F:\vlado\models\Stable-diffusion\novaAnimeXL_ponyV40.safetensors"
2024-10-16 14:25:35,895 | sd | INFO | sd_models | Autodetect model: detect="Stable Diffusion XL" class=StableDiffusionXLPipeline file="F:\vlado\models\Stable-diffusion\novaAnimeXL_ponyV40.safetensors" size=6617MB
2024-10-16 14:25:35,896 | sd | INFO | sd_models | Autodetect vae: detect="Stable Diffusion XL" class=StableDiffusionXLPipeline file="F:\vlado\models\Stable-diffusion\novaAnimeXL_ponyV40.safetensors" size=6617MB
2024-10-16 14:25:35,898 | sd | INFO | sd_vae | Load module: type=VAE model="C:\Users\zaxof\OneDrive\Documents\GitHub\nvidia_sdnext\models\VAE\fixFP16ErrorsSDXLLowerMemoryUse_v10.safetensors" source=settings config={'low_cpu_mem_usage': False, 'torch_dtype': torch.bfloat16, 'use_safetensors': True, 'variant': 'fp16', 'force_upcast': False}
2024-10-16 14:25:40,606 | sd | DEBUG | sd_models | Setting model: pipeline=StableDiffusionXLPipeline config={'low_cpu_mem_usage': True, 'torch_dtype': torch.bfloat16, 'load_connected_pipeline': True, 'variant': 'fp16', 'extract_ema': False, 'use_safetensors': True, 'cache_dir': 'C:\\Users\\zaxof\\.cache\\huggingface\\hub', 'force_upcast': False}
2024-10-16 14:25:47,019 | sd | INFO | textual_inversion | Load network: type=embeddings loaded=2 skipped=12 time=6.41
2024-10-16 14:25:47,021 | sd | DEBUG | sd_models | Setting model: component=VAE name="fixFP16ErrorsSDXLLowerMemoryUse_v10.safetensors"
2024-10-16 14:25:47,022 | sd | DEBUG | sd_models | Setting model: component=VAE upcast=False
2024-10-16 14:25:47,023 | sd | DEBUG | sd_models | Setting model: component=VAE tiling=True
2024-10-16 14:25:47,025 | sd | DEBUG | sd_models | Setting model: attention="Scaled-Dot-Product"
2024-10-16 14:25:47,045 | sd | DEBUG | sd_models | Setting model: offload=none
2024-10-16 14:25:47,047 | sd | DEBUG | shared | Read: file="C:\Users\zaxof\OneDrive\Documents\GitHub\nvidia_sdnext\configs\sdxl\vae\config.json" json=15 bytes=674 time=0.001
2024-10-16 14:25:50,179 | sd | DEBUG | devices | GC: utilization={'gpu': 65, 'ram': 7, 'threshold': 90} gc={'collected': 3457, 'saved': 0.0} before={'gpu': 7.82, 'ram': 4.25} after={'gpu': 7.82, 'ram': 4.25, 'retries': 0, 'oom': 0} device=cuda fn=reload_model_weights:load_diffuser time=0.21
2024-10-16 14:25:50,188 | sd | INFO | sd_models | Load model: time=14.07 load=4.72 embeddings=6.41 move=2.92 native=512 memory={'ram': {'used': 4.25, 'total': 63.9}, 'gpu': {'used': 7.82, 'total': 12.0}, 'retries': 0, 'oom': 0}
2024-10-16 14:25:50,210 | sd | INFO | processing_diffusers | Base: class=StableDiffusionXLPipeline
2024-10-16 14:25:50,212 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM++ 2M" class="DPMSolverMultistepScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'thresholding': False, 'sample_max_value': 1.0, 'algorithm_type': 'dpmsolver++', 'solver_type': 'midpoint', 'lower_order_final': True, 'use_karras_sigmas': True, 'final_sigmas_type': 'zero', 'timestep_spacing': 'leading', 'solver_order': 2}
2024-10-16 14:25:50,556 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[2443316845]
2024-10-16 14:25:50,558 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=1/1x1 set={'prompt_embeds': torch.Size([1, 77, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 77, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 6, 'num_inference_steps': 20, 'eta': 1.0, 'guidance_rescale': 0.7, 'denoising_end': None, 'output_type': 'latent', 'width': 1024, 'height': 1024, 'parser': 'Full parser'}
2024-10-16 14:26:00,000 | sd | DEBUG | launch | Server: alive=True jobs=1 requests=137 uptime=174 memory=3.67/63.9 backend=Backend.DIFFUSERS state=job="txt2img" 0/-1
2024-10-16 14:26:08,473 | sd | DEBUG | processing_vae | VAE decode: vae name="fixFP16ErrorsSDXLLowerMemoryUse_v10" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=False tiling=True latents shape=torch.Size([1, 4, 128, 128]) dtype=torch.bfloat16 device=cuda:0 time=3.697
2024-10-16 14:26:08,976 | sd | INFO | images | Save: image="F:\vlado\outputs\01051-novaAnimeXL_ponyV40-rating safe score 9 a car.png" type=PNG width=1024 height=1024 size=1388493
2024-10-16 14:26:09,014 | sd | INFO | processing | Processed: images=1 its=1.06 time=18.79 timers={'encode': 0.34, 'args': 0.38, 'callback': 5.68, 'pipeline': 14.16, 'decode': 3.77, 'post': 0.49} memory={'ram': {'used': 2.9, 'total': 63.9}, 'gpu': {'used': 7.9, 'total': 12.0}, 'retries': 0, 'oom': 0}
2024-10-16 14:26:57,903 | sd | DEBUG | networks | LoRA name="PonyMinimalStyle" type={'ModuleTypeLora'} keys=986
2024-10-16 14:26:57,933 | sd | INFO | extra_networks_lora | Load network: type=LoRA apply=['PonyMinimalStyle'] patch=0.00 te=[1.5] unet=[[1.5, 1.5, 1.5]] dims=[None] load=1.63
2024-10-16 14:26:57,936 | sd | INFO | processing_diffusers | Base: class=StableDiffusionXLPipeline
2024-10-16 14:26:57,938 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM++ 2M" class="DPMSolverMultistepScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'thresholding': False, 'sample_max_value': 1.0, 'algorithm_type': 'dpmsolver++', 'solver_type': 'midpoint', 'lower_order_final': True, 'use_karras_sigmas': True, 'final_sigmas_type': 'zero', 'timestep_spacing': 'leading', 'solver_order': 2}
2024-10-16 14:26:59,426 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[3287572709]
2024-10-16 14:26:59,427 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=1/1x1 set={'prompt_embeds': torch.Size([1, 77, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 77, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 6, 'num_inference_steps': 20, 'eta': 1.0, 'guidance_rescale': 0.7, 'denoising_end': None, 'output_type': 'latent', 'width': 1024, 'height': 1024, 'parser': 'Full parser'}
2024-10-16 14:27:16,707 | sd | DEBUG | processing_vae | VAE decode: vae name="fixFP16ErrorsSDXLLowerMemoryUse_v10" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=False tiling=True latents shape=torch.Size([1, 4, 128, 128]) dtype=torch.bfloat16 device=cuda:0 time=1.081
2024-10-16 14:27:17,055 | sd | INFO | images | Save: image="F:\vlado\outputs\01052-novaAnimeXL_ponyV40-rating safe score 9 a car PonyMinimalStyle.png" type=PNG width=1024 height=1024 size=739589
2024-10-16 14:27:17,092 | sd | INFO | processing | Processed: images=1 its=0.96 time=20.77 timers={'init': 1.65, 'encode': 1.49, 'args': 1.5, 'callback': 5.63, 'pipeline': 16.17, 'decode': 1.1, 'post': 0.38} memory={'ram': {'used': 8.31, 'total': 63.9}, 'gpu': {'used': 9.61, 'total': 12.0}, 'retries': 0, 'oom': 0}
2024-10-16 14:27:45,722 | sd | INFO | processing_diffusers | Base: class=StableDiffusionXLPipeline
2024-10-16 14:27:45,723 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM++ 2M" class="DPMSolverMultistepScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'thresholding': False, 'sample_max_value': 1.0, 'algorithm_type': 'dpmsolver++', 'solver_type': 'midpoint', 'lower_order_final': True, 'use_karras_sigmas': True, 'final_sigmas_type': 'zero', 'timestep_spacing': 'leading', 'solver_order': 2}
2024-10-16 14:27:46,071 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[921354225]
2024-10-16 14:27:46,073 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=1/1x1 set={'prompt_embeds': torch.Size([1, 77, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 77, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 6, 'num_inference_steps': 20, 'eta': 1.0, 'guidance_rescale': 0.7, 'denoising_end': None, 'output_type': 'latent', 'width': 1024, 'height': 1024, 'parser': 'Full parser'}
2024-10-16 14:27:59,503 | sd | DEBUG | processing_vae | VAE decode: vae name="fixFP16ErrorsSDXLLowerMemoryUse_v10" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=False tiling=True latents shape=torch.Size([1, 4, 128, 128]) dtype=torch.bfloat16 device=cuda:0 time=1.079
2024-10-16 14:27:59,917 | sd | DEBUG | launch | Server: alive=True jobs=1 requests=231 uptime=293 memory=8.31/63.9 backend=Backend.DIFFUSERS state=job="txt2img" 0/-1
2024-10-16 14:27:59,942 | sd | INFO | images | Save: image="F:\vlado\outputs\01053-novaAnimeXL_ponyV40-rating safe score 9 a car.png" type=PNG width=1024 height=1024 size=941135
2024-10-16 14:27:59,978 | sd | INFO | processing | Processed: images=1 its=1.41 time=14.23 timers={'encode': 0.35, 'args': 0.36, 'callback': 5.62, 'pipeline': 12.32, 'decode': 1.1, 'post': 0.47} memory={'ram': {'used': 8.31, 'total': 63.9}, 'gpu': {'used': 9.61, 'total': 12.0}, 'retries': 0, 'oom': 0}
2024-10-16 14:29:59,827 | sd | DEBUG | launch | Server: alive=True jobs=1 requests=252 uptime=413 memory=8.29/63.9 backend=Backend.DIFFUSERS state=job="txt2img" 0/-1
2024-10-16 14:31:59,739 | sd | DEBUG | launch | Server: alive=True jobs=1 requests=266 uptime=533 memory=8.29/63.9 backend=Backend.DIFFUSERS state=job="txt2img" 0/-1
2024-10-16 14:33:59,654 | sd | DEBUG | launch | Server: alive=True jobs=1 requests=290 uptime=653 memory=8.29/63.9 backend=Backend.DIFFUSERS state=job="txt2img" 0/-1
2024-10-16 14:35:59,569 | sd | DEBUG | launch | Server: alive=True jobs=1 requests=314 uptime=773 memory=8.29/63.9 backend=Backend.DIFFUSERS state=job="txt2img" 0/-1
2024-10-16 14:38:00,483 | sd | DEBUG | launch | Server: alive=True jobs=1 requests=337 uptime=894 memory=8.29/63.9 backend=Backend.DIFFUSERS state=job="txt2img" 0/-1
2024-10-16 14:40:00,394 | sd | DEBUG | launch | Server: alive=True jobs=1 requests=361 uptime=1014 memory=8.3/63.9 backend=Backend.DIFFUSERS state=job="txt2img" 0/-1

Backend

Diffusers

UI

Standard

Branch

Dev

Model

StableDiffusion XL

Acknowledgements

vladmandic commented 4 days ago

would be great if you could reproduce with env variable SD_LORA_DEBUG=true set

zaxwashere commented 4 days ago

sdnext log lora_debug.txt

no lora no lora lora on lora on lora removed lora removed

added the env variable, ran it again with --safe to disable all extensions. Let me know if you need anything else

vladmandic commented 3 days ago

i'm surprised this was not reported long time ago since this logic bug in the code existed for a while. basically, if there were no loras required, deactivate function disabled patched methods so they would not actually execute to restore weights from backup. anyhow, fixed.