[Issue]: New X/Y/Z grid has unexpected behavior when batch count > 1

Issue Description

NOTE: Not sure if this should be an extension issue, the issue is with the new "built-in" X/Y/Z grid. Does that count as core or extension?

When generating a X/Y/Z grid, when it is done it addes duplicates (same as the batch count) of the end result. With larger grids it makes the UI extremly slow, I assume it loads them into the browser.

Version Platform Description

2024-10-27 16:50:22,730 | sd | INFO | launch | Starting SD.Next 2024-10-27 16:50:22,732 | sd | INFO | installer | Logger: file="e:\sdnext\sdnext.log" level=INFO size=65 mode=append 2024-10-27 16:50:22,734 | sd | INFO | installer | Python: version=3.10.6 platform=Windows bin="e:\sdnext\venv\Scripts\python.exe" venv="e:\sdnext\venv" 2024-10-27 16:50:22,831 | sd | INFO | installer | Version: app=sd.next updated=2024-10-24 hash=06dd9e5fc branch=master url=https://github.com/vladmandic/automatic.git/tree/master ui=main 2024-10-27 16:50:23,145 | sd | INFO | launch | Platform: arch=AMD64 cpu=Intel64 Family 6 Model 183 Stepping 1, GenuineIntel system=Windows release=Windows-10-10.0.22631-SP0 python=3.10.6 2024-10-27 16:50:23,147 | sd | DEBUG | installer | Setting environment tuning 2024-10-27 16:50:23,157 | sd | DEBUG | installer | Torch allocator: "garbage_collection_threshold:0.80,max_split_size_mb:512" 2024-10-27 16:50:23,172 | sd | DEBUG | installer | Torch overrides: cuda=False rocm=False ipex=False directml=False openvino=False zluda=False 2024-10-27 16:50:23,177 | sd | INFO | installer | CUDA: nVidia toolkit detected 2024-10-27 16:50:23,644 | sd | INFO | installer | Verifying requirements 2024-10-27 16:50:23,646 | sd | INFO | installer | Verifying packages 2024-10-27 16:50:23,663 | sd | DEBUG | installer | Timestamp repository update time: Fri Oct 25 01:48:38 2024 2024-10-27 16:50:23,663 | sd | INFO | launch | Startup: standard 2024-10-27 16:50:23,664 | sd | INFO | installer | Verifying submodules 2024-10-27 16:50:24,961 | sd | DEBUG | installer | Git submodule: extensions-builtin/sd-extension-chainner / main 2024-10-27 16:50:24,994 | sd | DEBUG | installer | Git submodule: extensions-builtin/sd-extension-system-info / main 2024-10-27 16:50:25,023 | sd | DEBUG | installer | Git submodule: extensions-builtin/sd-webui-agent-scheduler / main 2024-10-27 16:50:25,067 | sd | DEBUG | installer | Git detached head detected: folder="extensions-builtin/sdnext-modernui" reattach=dev 2024-10-27 16:50:25,067 | sd | DEBUG | installer | Git submodule: extensions-builtin/sdnext-modernui / dev 2024-10-27 16:50:25,101 | sd | DEBUG | installer | Git submodule: extensions-builtin/stable-diffusion-webui-rembg / master 2024-10-27 16:50:25,131 | sd | DEBUG | installer | Git submodule: modules/k-diffusion / master 2024-10-27 16:50:25,162 | sd | DEBUG | installer | Git submodule: wiki / master 2024-10-27 16:50:25,181 | sd | DEBUG | paths | Register paths 2024-10-27 16:50:25,214 | sd | DEBUG | installer | Installed packages: 196 2024-10-27 16:50:25,214 | sd | DEBUG | installer | Extensions all: ['Lora', 'sd-extension-chainner', 'sd-extension-system-info', 'sd-webui-agent-scheduler', 'stable-diffusion-webui-rembg'] 2024-10-27 16:50:25,298 | sd | DEBUG | installer | Extension installer: E:\sdnext\extensions-builtin\sd-webui-agent-scheduler\install.py 2024-10-27 16:50:26,720 | sd | DEBUG | installer | Extension installer: E:\sdnext\extensions-builtin\stable-diffusion-webui-rembg\install.py 2024-10-27 16:50:30,739 | sd | DEBUG | installer | Extensions all: ['adetailer', 'sd-dynamic-prompts', 'stable-diffusion-webui-state'] 2024-10-27 16:50:30,739 | sd | DEBUG | installer | Extension force: name="adetailer" commit=a89c01d 2024-10-27 16:50:30,764 | sd | DEBUG | installer | Extension installer: E:\sdnext\extensions\adetailer\install.py 2024-10-27 16:50:30,852 | sd | DEBUG | installer | Extension installer: E:\sdnext\extensions\sd-dynamic-prompts\install.py 2024-10-27 16:50:31,054 | sd | INFO | installer | Extensions enabled: ['Lora', 'sd-extension-chainner', 'sd-extension-system-info', 'sd-webui-agent-scheduler', 'stable-diffusion-webui-rembg', 'adetailer', 'sd-dynamic-prompts', 'stable-diffusion-webui-state'] 2024-10-27 16:50:31,056 | sd | INFO | installer | Verifying requirements 2024-10-27 16:50:31,057 | sd | DEBUG | launch | Setup complete without errors: 1730044231 2024-10-27 16:50:31,063 | sd | DEBUG | installer | Extension preload: {'extensions-builtin': 0.0, 'extensions': 0.0} 2024-10-27 16:50:31,064 | sd | DEBUG | launch | Starting module: <module 'webui' from 'e:\sdnext\webui.py'> 2024-10-27 16:50:31,064 | sd | INFO | launch | Command line args: [] 2024-10-27 16:50:31,065 | sd | DEBUG | launch | Env flags: [] 2024-10-27 16:50:39,253 | sd | INFO | loader | System packages: {'torch': '2.4.1+cu124', 'diffusers': '0.32.0.dev0', 'gradio': '3.43.2', 'transformers': '4.45.2', 'accelerate': '1.0.0'} 2024-10-27 16:50:39,875 | sd | DEBUG | shared | Huggingface cache: folder="C:\Users\Seafan.cache\huggingface\hub" 2024-10-27 16:50:39,962 | sd | INFO | shared | Device detect: memory=24.0 optimization=none 2024-10-27 16:50:39,965 | sd | DEBUG | shared | Read: file="config.json" json=53 bytes=3597 time=0.000 2024-10-27 16:50:39,965 | sd | DEBUG | shared | Unknown settings: ['state', 'state_txt2img', 'state_img2img', 'state_extensions', 'dp_ignore_whitespace', 'dp_write_raw_template', 'multiple_tqdm', 'ldsr_steps', 'facehires_strength', 'facehires_conf', 'facehires_max'] 2024-10-27 16:50:39,965 | sd | INFO | shared | Engine: backend=Backend.DIFFUSERS compute=cuda device=cuda attention="Scaled-Dot-Product" mode=no_grad 2024-10-27 16:50:39,967 | sd | DEBUG | shared | Read: file="html\reference.json" json=58 bytes=30996 time=0.000 2024-10-27 16:50:40,028 | 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" 2024-10-27 16:50:43,209 | sd | ERROR | init | ONNX failed to initialize: DLL load failed while importing onnx_cpp2py_export: A dynamic link library (DLL) initialization routine failed. 2024-10-27 16:50:46,797 | sd | INFO | shared | Device: device=NVIDIA GeForce RTX 4090 n=1 arch=sm_90 capability=(8, 9) cuda=12.4 cudnn=90100 driver=565.90 2024-10-27 16:50:46,933 | sd | DEBUG | sd_hijack | Importing LDM

Relevant log output

2024-10-27 16:51:00,820 | sd | INFO | textual_inversion | Load network: type=embeddings loaded=0 skipped=0 time=0.00
2024-10-27 16:51:00,822 | sd | DEBUG | sd_models | Setting model: component=VAE slicing=True
2024-10-27 16:51:00,822 | sd | DEBUG | sd_models | Setting model: component=VAE tiling=True
2024-10-27 16:51:00,822 | sd | DEBUG | sd_models | Setting model: attention="Scaled-Dot-Product"
2024-10-27 16:51:00,844 | sd | DEBUG | sd_models | Setting model: offload=none
2024-10-27 16:51:00,844 | sd | DEBUG | shared | Read: file="E:\sdnext\configs\sdxl\vae\config.json" json=15 bytes=674 time=0.000
2024-10-27 16:51:02,452 | sd | DEBUG | devices | GC: utilization={'gpu': 35, 'ram': 12, 'threshold': 80} gc={'collected': 254, 'saved': 0.0} before={'gpu': 8.37, 'ram': 7.91} after={'gpu': 8.37, 'ram': 7.91, 'retries': 0, 'oom': 0} device=cuda fn=reload_model_weights:load_diffuser time=0.3
2024-10-27 16:51:02,462 | sd | INFO | sd_models | Load model: time=8.71 load=7.38 move=1.30 native=1024 memory={'ram': {'used': 7.91, 'total': 63.75}, 'gpu': {'used': 8.37, 'total': 23.99}, 'retries': 0, 'oom': 0}
2024-10-27 16:51:02,465 | sd | DEBUG | script_callbacks | Script callback init time: system-info.py:app_started=0.09 task_scheduler.py:app_started=0.21
2024-10-27 16:51:02,465 | sd | INFO | webui | Startup time: 31.40 torch=6.96 gradio=0.96 diffusers=0.10 libraries=7.82 extensions=1.42 models=0.09 detailer=0.10 ui-networks=1.68 ui-txt2img=0.46 ui-img2img=0.21 ui-control=0.29 ui-extras=0.05 ui-models=0.06 ui-settings=0.49 ui-extensions=0.59 ui-defaults=0.07 launch=0.49 api=0.08 app-started=0.30 checkpoint=9.03
2024-10-27 16:51:02,472 | sd | DEBUG | shared | Save: file="config.json" json=53 bytes=3457 time=0.007
2024-10-27 16:51:02,472 | sd | DEBUG | shared | Unused settings: ['multiple_tqdm', 'facehires_strength', 'facehires_conf', 'facehires_max']
2024-10-27 16:51:25,311 | sd | INFO | xyz_grid_on | XYZ grid start: images=9 grid=1 shape=1x3 cells=1 steps=405
2024-10-27 16:51:25,348 | sd | INFO | processing_diffusers | Base: class=StableDiffusionXLPipeline
2024-10-27 16:51:25,384 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:51:25,990 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393359]
2024-10-27 16:51:25,991 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=1/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 4, 'num_inference_steps': 30, 'eta': 1.0, 'guidance_rescale': 0.7, 'denoising_end': None, 'output_type': 'latent', 'width': 512, 'height': 512, 'parser': 'Full parser'}
2024-10-27 16:51:27,542 | sd | DEBUG | sd_vae_approx | VAE load: type=approximate model=models\VAE-approx\model.pt
2024-10-27 16:51:30,764 | sd | DEBUG | processing_class | Init hires: upscaler="Aura SR 4x" sampler="DPM SDE" resize=0x0 upscale=1126x1126
2024-10-27 16:51:30,764 | sd | INFO | processing_diffusers | Upscale: mode=1 upscaler="Aura SR 4x" context="Add with forward" resize=0x0 upscale=1126x1126
2024-10-27 16:51:31,984 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 64, 64]) dtype=torch.bfloat16 device=cuda:0 time=1.219
2024-10-27 16:51:42,116 | sd | DEBUG | images_resize | Image resize: input=<PIL.Image.Image image mode=RGB size=512x512 at 0x2B638CE9ED0> width=1126 height=1126 mode="Fixed" upscaler="Aura SR 4x" context="Add with forward" type=image result=<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B6383B6F50> time=10.09 fn=process_hires:resize_hires
2024-10-27 16:51:42,117 | sd | DEBUG | sd_models | Pipeline class change: original=StableDiffusionXLPipeline target=StableDiffusionXLImg2ImgPipeline device=cuda:0 fn=process_diffusers:process_hires
2024-10-27 16:51:42,117 | sd | INFO | processing_diffusers | HiRes: class=StableDiffusionXLImg2ImgPipeline sampler="DPM SDE"
2024-10-27 16:51:42,135 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:51:42,291 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393359]
2024-10-27 16:51:42,291 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLImg2ImgPipeline task=DiffusersTaskType.IMAGE_2_IMAGE batch=1/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 6, 'num_inference_steps': 51, 'eta': 1.0, 'guidance_rescale': 0.7, 'output_type': 'latent', 'image': [<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B6383B6F50>], 'strength': 0.3, 'parser': 'Full parser'}
2024-10-27 16:51:57,635 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 140, 140]) dtype=torch.bfloat16 device=cuda:0 time=5.584
2024-10-27 16:51:57,670 | sd | INFO | images | Save: image="outputs\text\51716-tPonynai3_v6_DPM SDE_1986393359_4_30_score 9 score 8 up score 7 up.jpg" type=JPEG width=1120 height=1120 size=203737
2024-10-27 16:51:57,688 | sd | INFO | processing_diffusers | Base: class=StableDiffusionXLPipeline
2024-10-27 16:51:57,704 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:51:57,839 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393360]
2024-10-27 16:51:57,839 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=2/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 4, 'num_inference_steps': 30, 'eta': 1.0, 'guidance_rescale': 0.7, 'denoising_end': None, 'output_type': 'latent', 'width': 512, 'height': 512, 'parser': 'Full parser'}
2024-10-27 16:51:59,932 | sd | DEBUG | launch | Server: alive=True jobs=1 requests=80 uptime=80 memory=5.01/63.75 backend=Backend.DIFFUSERS state=idle
2024-10-27 16:52:01,536 | sd | DEBUG | processing_class | Init hires: upscaler="Aura SR 4x" sampler="DPM SDE" resize=0x0 upscale=1126x1126
2024-10-27 16:52:01,536 | sd | INFO | processing_diffusers | Upscale: mode=1 upscaler="Aura SR 4x" context="Add with forward" resize=0x0 upscale=1126x1126
2024-10-27 16:52:01,545 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 64, 64]) dtype=torch.bfloat16 device=cuda:0 time=0.008
2024-10-27 16:52:03,813 | sd | DEBUG | images_resize | Image resize: input=<PIL.Image.Image image mode=RGB size=512x512 at 0x2B8ADA4E950> width=1126 height=1126 mode="Fixed" upscaler="Aura SR 4x" context="Add with forward" type=image result=<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B63832FB50> time=2.24 fn=process_hires:resize_hires
2024-10-27 16:52:03,814 | sd | DEBUG | sd_models | Pipeline class change: original=StableDiffusionXLPipeline target=StableDiffusionXLImg2ImgPipeline device=cuda:0 fn=process_diffusers:process_hires
2024-10-27 16:52:03,814 | sd | INFO | processing_diffusers | HiRes: class=StableDiffusionXLImg2ImgPipeline sampler="DPM SDE"
2024-10-27 16:52:03,832 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:52:03,968 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393360]
2024-10-27 16:52:03,968 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLImg2ImgPipeline task=DiffusersTaskType.IMAGE_2_IMAGE batch=2/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 6, 'num_inference_steps': 51, 'eta': 1.0, 'guidance_rescale': 0.7, 'output_type': 'latent', 'image': [<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B63832FB50>], 'strength': 0.3, 'parser': 'Full parser'}
2024-10-27 16:52:09,138 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 140, 140]) dtype=torch.bfloat16 device=cuda:0 time=0.349
2024-10-27 16:52:09,383 | sd | DEBUG | devices | GC: utilization={'gpu': 86, 'ram': 8, 'threshold': 80} gc={'collected': 2464, 'saved': 12.11} before={'gpu': 20.61, 'ram': 5.04} after={'gpu': 8.5, 'ram': 5.04, 'retries': 0, 'oom': 0} device=cuda fn=process_decode:vae_decode time=0.24
2024-10-27 16:52:09,408 | sd | INFO | images | Save: image="outputs\text\51717-tPonynai3_v6_DPM SDE_1986393360_4_30_score 9 score 8 up score 7 up.jpg" type=JPEG width=1120 height=1120 size=174966
2024-10-27 16:52:09,421 | sd | INFO | processing_diffusers | Base: class=StableDiffusionXLPipeline
2024-10-27 16:52:09,436 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:52:09,576 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393361]
2024-10-27 16:52:09,576 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=3/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 4, 'num_inference_steps': 30, 'eta': 1.0, 'guidance_rescale': 0.7, 'denoising_end': None, 'output_type': 'latent', 'width': 512, 'height': 512, 'parser': 'Full parser'}
2024-10-27 16:52:13,518 | sd | DEBUG | processing_class | Init hires: upscaler="Aura SR 4x" sampler="DPM SDE" resize=0x0 upscale=1126x1126
2024-10-27 16:52:13,519 | sd | INFO | processing_diffusers | Upscale: mode=1 upscaler="Aura SR 4x" context="Add with forward" resize=0x0 upscale=1126x1126
2024-10-27 16:52:13,529 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 64, 64]) dtype=torch.bfloat16 device=cuda:0 time=0.009
2024-10-27 16:52:15,773 | sd | DEBUG | images_resize | Image resize: input=<PIL.Image.Image image mode=RGB size=512x512 at 0x2B8ADA4C520> width=1126 height=1126 mode="Fixed" upscaler="Aura SR 4x" context="Add with forward" type=image result=<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B638CA6170> time=2.21 fn=process_hires:resize_hires
2024-10-27 16:52:15,774 | sd | DEBUG | sd_models | Pipeline class change: original=StableDiffusionXLPipeline target=StableDiffusionXLImg2ImgPipeline device=cuda:0 fn=process_diffusers:process_hires
2024-10-27 16:52:15,775 | sd | INFO | processing_diffusers | HiRes: class=StableDiffusionXLImg2ImgPipeline sampler="DPM SDE"
2024-10-27 16:52:15,791 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:52:15,928 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393361]
2024-10-27 16:52:15,928 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLImg2ImgPipeline task=DiffusersTaskType.IMAGE_2_IMAGE batch=3/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 6, 'num_inference_steps': 51, 'eta': 1.0, 'guidance_rescale': 0.7, 'output_type': 'latent', 'image': [<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B638CA6170>], 'strength': 0.3, 'parser': 'Full parser'}
2024-10-27 16:52:21,047 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 140, 140]) dtype=torch.bfloat16 device=cuda:0 time=0.328
2024-10-27 16:52:21,281 | sd | DEBUG | devices | GC: utilization={'gpu': 86, 'ram': 8, 'threshold': 80} gc={'collected': 3337, 'saved': 11.97} before={'gpu': 20.52, 'ram': 5.04} after={'gpu': 8.55, 'ram': 5.04, 'retries': 0, 'oom': 0} device=cuda fn=process_decode:vae_decode time=0.23
2024-10-27 16:52:21,305 | sd | INFO | images | Save: image="outputs\text\51718-tPonynai3_v6_DPM SDE_1986393361_4_30_score 9 score 8 up score 7 up.jpg" type=JPEG width=1120 height=1120 size=214174
2024-10-27 16:52:21,372 | sd | INFO | images | Save: image="outputs\grids\18392-tPonynai3_v6_DPM SDE_1986393359_4_30_score 9 score 8 up score 7 up-grid.jpg" type=JPEG width=3360 height=1120 size=589565
2024-10-27 16:52:21,382 | sd | INFO | processing | Processed: images=4 its=2.14 time=55.98 timers={'gc': 0.48, 'init': 0.05, 'prepare': 0.08, 'encode': 17.3, 'args': 0.97, 'move': 0.02, 'callback': 4.78, 'pipeline': 12.24, 'hires': 35.81, 'decode': 6.75, 'post': 0.16} memory={'ram': {'used': 5.04, 'total': 63.75}, 'gpu': {'used': 8.55, 'total': 23.99}, 'retries': 0, 'oom': 0}
2024-10-27 16:52:27,171 | sd | DEBUG | devices | GC: utilization={'gpu': 36, 'ram': 18, 'threshold': 80} gc={'collected': 1406, 'saved': 6.87} before={'gpu': 8.55, 'ram': 11.38} after={'gpu': 1.68, 'ram': 11.38, 'retries': 0, 'oom': 0} device=cuda fn=reload_model_weights:unload_model_weights time=0.23
2024-10-27 16:52:27,172 | sd | DEBUG | sd_models | Unload weights model: {'ram': {'used': 11.38, 'total': 63.75}, 'gpu': {'used': 1.68, 'total': 23.99}, 'retries': 0, 'oom': 0}
2024-10-27 16:52:27,172 | sd | INFO | sd_models | Autodetect model: detect="Stable Diffusion XL" class=StableDiffusionXLPipeline file="d:\StableDiffusion\models\pony\anime\veridian_.safetensors" size=6617MB
2024-10-27 16:52:27,173 | sd | DEBUG | sd_models | Diffusers accelerate: direct=True
2024-10-27 16:52:32,510 | sd | DEBUG | sd_models | Setting model: pipeline=StableDiffusionXLPipeline config={'low_cpu_mem_usage': True, 'torch_dtype': torch.bfloat16, 'load_connected_pipeline': True, 'extract_ema': False, 'config': 'configs/sdxl', 'use_safetensors': True, 'cache_dir': 'C:\\Users\\Seafan\\.cache\\huggingface\\hub'}
2024-10-27 16:52:32,511 | sd | INFO | textual_inversion | Load network: type=embeddings loaded=0 skipped=0 time=0.00
2024-10-27 16:52:32,511 | sd | DEBUG | sd_models | Setting model: component=VAE slicing=True
2024-10-27 16:52:32,512 | sd | DEBUG | sd_models | Setting model: component=VAE tiling=True
2024-10-27 16:52:32,512 | sd | DEBUG | sd_models | Setting model: attention="Scaled-Dot-Product"
2024-10-27 16:52:32,526 | sd | DEBUG | sd_models | Setting model: offload=none
2024-10-27 16:52:32,527 | sd | DEBUG | shared | Read: file="E:\sdnext\configs\sdxl\vae\config.json" json=15 bytes=674 time=0.000
2024-10-27 16:52:33,510 | sd | DEBUG | devices | GC: utilization={'gpu': 35, 'ram': 19, 'threshold': 80} gc={'collected': 450, 'saved': 0.0} before={'gpu': 8.42, 'ram': 12.13} after={'gpu': 8.42, 'ram': 12.13, 'retries': 0, 'oom': 0} device=cuda fn=reload_model_weights:load_diffuser time=0.19
2024-10-27 16:52:33,514 | sd | INFO | sd_models | Load model: time=6.15 load=5.34 move=0.80 native=1024 memory={'ram': {'used': 12.13, 'total': 63.75}, 'gpu': {'used': 8.42, 'total': 23.99}, 'retries': 0, 'oom': 0}
2024-10-27 16:52:33,517 | sd | DEBUG | xyz_grid_shared | XYZ grid apply checkpoint: "pony\anime\veridian_ [d25c59fa91]"
2024-10-27 16:52:33,517 | sd | DEBUG | processing | Override: {'sd_model_checkpoint': 'pony\\anime\\veridian_'}
2024-10-27 16:52:33,517 | sd | INFO | sd_models | Load model: select="pony\anime\veridian_ [d25c59fa91]"
2024-10-27 16:52:33,535 | sd | INFO | processing_diffusers | Base: class=StableDiffusionXLPipeline
2024-10-27 16:52:33,550 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:52:33,698 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393359]
2024-10-27 16:52:33,698 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=1/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 4, 'num_inference_steps': 30, 'eta': 1.0, 'guidance_rescale': 0.7, 'denoising_end': None, 'output_type': 'latent', 'width': 512, 'height': 512, 'parser': 'Full parser'}
2024-10-27 16:52:37,362 | sd | DEBUG | processing_class | Init hires: upscaler="Aura SR 4x" sampler="DPM SDE" resize=0x0 upscale=1126x1126
2024-10-27 16:52:37,362 | sd | INFO | processing_diffusers | Upscale: mode=1 upscaler="Aura SR 4x" context="Add with forward" resize=0x0 upscale=1126x1126
2024-10-27 16:52:37,371 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 64, 64]) dtype=torch.bfloat16 device=cuda:0 time=0.008
2024-10-27 16:52:39,820 | sd | DEBUG | images_resize | Image resize: input=<PIL.Image.Image image mode=RGB size=512x512 at 0x2B63B278370> width=1126 height=1126 mode="Fixed" upscaler="Aura SR 4x" context="Add with forward" type=image result=<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B6390D5480> time=2.41 fn=process_hires:resize_hires
2024-10-27 16:52:39,822 | sd | DEBUG | sd_models | Pipeline class change: original=StableDiffusionXLPipeline target=StableDiffusionXLImg2ImgPipeline device=cuda:0 fn=process_diffusers:process_hires
2024-10-27 16:52:39,822 | sd | INFO | processing_diffusers | HiRes: class=StableDiffusionXLImg2ImgPipeline sampler="DPM SDE"
2024-10-27 16:52:39,839 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:52:40,011 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393359]
2024-10-27 16:52:40,011 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLImg2ImgPipeline task=DiffusersTaskType.IMAGE_2_IMAGE batch=1/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 6, 'num_inference_steps': 51, 'eta': 1.0, 'guidance_rescale': 0.7, 'output_type': 'latent', 'image': [<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B6390D5480>], 'strength': 0.3, 'parser': 'Full parser'}
2024-10-27 16:52:45,136 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 140, 140]) dtype=torch.bfloat16 device=cuda:0 time=0.337
2024-10-27 16:52:45,391 | sd | DEBUG | devices | GC: utilization={'gpu': 86, 'ram': 8, 'threshold': 80} gc={'collected': 2897, 'saved': 11.99} before={'gpu': 20.53, 'ram': 5.08} after={'gpu': 8.54, 'ram': 5.04, 'retries': 0, 'oom': 0} device=cuda fn=process_decode:vae_decode time=0.25
2024-10-27 16:52:45,413 | sd | INFO | images | Save: image="outputs\text\51719-veridian__DPM SDE_1986393359_4_30_score 9 score 8 up score 7 up.jpg" type=JPEG width=1120 height=1120 size=165684
2024-10-27 16:52:45,427 | sd | INFO | processing_diffusers | Base: class=StableDiffusionXLPipeline
2024-10-27 16:52:45,444 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:52:45,582 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393360]
2024-10-27 16:52:45,582 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=2/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 4, 'num_inference_steps': 30, 'eta': 1.0, 'guidance_rescale': 0.7, 'denoising_end': None, 'output_type': 'latent', 'width': 512, 'height': 512, 'parser': 'Full parser'}
2024-10-27 16:52:49,547 | sd | DEBUG | processing_class | Init hires: upscaler="Aura SR 4x" sampler="DPM SDE" resize=0x0 upscale=1126x1126
2024-10-27 16:52:49,547 | sd | INFO | processing_diffusers | Upscale: mode=1 upscaler="Aura SR 4x" context="Add with forward" resize=0x0 upscale=1126x1126
2024-10-27 16:52:49,557 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 64, 64]) dtype=torch.bfloat16 device=cuda:0 time=0.009
2024-10-27 16:52:51,807 | sd | DEBUG | images_resize | Image resize: input=<PIL.Image.Image image mode=RGB size=512x512 at 0x2B63B27AAA0> width=1126 height=1126 mode="Fixed" upscaler="Aura SR 4x" context="Add with forward" type=image result=<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B6391687C0> time=2.22 fn=process_hires:resize_hires
2024-10-27 16:52:51,808 | sd | DEBUG | sd_models | Pipeline class change: original=StableDiffusionXLPipeline target=StableDiffusionXLImg2ImgPipeline device=cuda:0 fn=process_diffusers:process_hires
2024-10-27 16:52:51,808 | sd | INFO | processing_diffusers | HiRes: class=StableDiffusionXLImg2ImgPipeline sampler="DPM SDE"
2024-10-27 16:52:51,825 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:52:51,961 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393360]
2024-10-27 16:52:51,961 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLImg2ImgPipeline task=DiffusersTaskType.IMAGE_2_IMAGE batch=2/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 6, 'num_inference_steps': 51, 'eta': 1.0, 'guidance_rescale': 0.7, 'output_type': 'latent', 'image': [<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B6391687C0>], 'strength': 0.3, 'parser': 'Full parser'}
2024-10-27 16:52:57,195 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 140, 140]) dtype=torch.bfloat16 device=cuda:0 time=0.335
2024-10-27 16:52:57,445 | sd | DEBUG | devices | GC: utilization={'gpu': 86, 'ram': 8, 'threshold': 80} gc={'collected': 3276, 'saved': 12.17} before={'gpu': 20.69, 'ram': 5.07} after={'gpu': 8.52, 'ram': 5.07, 'retries': 0, 'oom': 0} device=cuda fn=process_decode:vae_decode time=0.25
2024-10-27 16:52:57,467 | sd | INFO | images | Save: image="outputs\text\51720-veridian__DPM SDE_1986393360_4_30_score 9 score 8 up score 7 up.jpg" type=JPEG width=1120 height=1120 size=164594
2024-10-27 16:52:57,481 | sd | INFO | processing_diffusers | Base: class=StableDiffusionXLPipeline
2024-10-27 16:52:57,497 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:52:57,629 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393361]
2024-10-27 16:52:57,630 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=3/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 4, 'num_inference_steps': 30, 'eta': 1.0, 'guidance_rescale': 0.7, 'denoising_end': None, 'output_type': 'latent', 'width': 512, 'height': 512, 'parser': 'Full parser'}
2024-10-27 16:53:01,476 | sd | DEBUG | processing_class | Init hires: upscaler="Aura SR 4x" sampler="DPM SDE" resize=0x0 upscale=1126x1126
2024-10-27 16:53:01,476 | sd | INFO | processing_diffusers | Upscale: mode=1 upscaler="Aura SR 4x" context="Add with forward" resize=0x0 upscale=1126x1126
2024-10-27 16:53:01,486 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 64, 64]) dtype=torch.bfloat16 device=cuda:0 time=0.009
2024-10-27 16:53:03,667 | sd | DEBUG | images_resize | Image resize: input=<PIL.Image.Image image mode=RGB size=512x512 at 0x2B63B278760> width=1126 height=1126 mode="Fixed" upscaler="Aura SR 4x" context="Add with forward" type=image result=<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B63913C670> time=2.15 fn=process_hires:resize_hires
2024-10-27 16:53:03,669 | sd | DEBUG | sd_models | Pipeline class change: original=StableDiffusionXLPipeline target=StableDiffusionXLImg2ImgPipeline device=cuda:0 fn=process_diffusers:process_hires
2024-10-27 16:53:03,669 | sd | INFO | processing_diffusers | HiRes: class=StableDiffusionXLImg2ImgPipeline sampler="DPM SDE"
2024-10-27 16:53:03,687 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:53:03,827 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393361]
2024-10-27 16:53:03,827 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLImg2ImgPipeline task=DiffusersTaskType.IMAGE_2_IMAGE batch=3/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 6, 'num_inference_steps': 51, 'eta': 1.0, 'guidance_rescale': 0.7, 'output_type': 'latent', 'image': [<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B63913C670>], 'strength': 0.3, 'parser': 'Full parser'}
2024-10-27 16:53:08,980 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 140, 140]) dtype=torch.bfloat16 device=cuda:0 time=0.328
2024-10-27 16:53:09,217 | sd | DEBUG | devices | GC: utilization={'gpu': 86, 'ram': 8, 'threshold': 80} gc={'collected': 3276, 'saved': 12.13} before={'gpu': 20.69, 'ram': 5.08} after={'gpu': 8.56, 'ram': 5.08, 'retries': 0, 'oom': 0} device=cuda fn=process_decode:vae_decode time=0.23
2024-10-27 16:53:09,249 | sd | INFO | images | Save: image="outputs\text\51721-veridian__DPM SDE_1986393361_4_30_score 9 score 8 up score 7 up.jpg" type=JPEG width=1120 height=1120 size=165400
2024-10-27 16:53:09,301 | sd | INFO | images | Save: image="outputs\grids\18393-veridian__DPM SDE_1986393359_4_30_score 9 score 8 up score 7 up-grid.jpg" type=JPEG width=3360 height=1120 size=489500
2024-10-27 16:53:09,310 | sd | INFO | processing | Processed: images=4 its=3.36 time=35.73 timers={'gc': 0.75, 'init': 0.02, 'prepare': 0.07, 'encode': 7.85, 'args': 0.43, 'callback': 4.91, 'pipeline': 11.42, 'hires': 21.95, 'decode': 1.74, 'post': 0.14} memory={'ram': {'used': 5.08, 'total': 63.75}, 'gpu': {'used': 8.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
2024-10-27 16:53:11,140 | sd | DEBUG | devices | GC: utilization={'gpu': 36, 'ram': 18, 'threshold': 80} gc={'collected': 1066, 'saved': 6.88} before={'gpu': 8.56, 'ram': 11.42} after={'gpu': 1.68, 'ram': 11.42, 'retries': 0, 'oom': 0} device=cuda fn=reload_model_weights:unload_model_weights time=0.22
2024-10-27 16:53:11,141 | sd | DEBUG | sd_models | Unload weights model: {'ram': {'used': 11.42, 'total': 63.75}, 'gpu': {'used': 1.68, 'total': 23.99}, 'retries': 0, 'oom': 0}
2024-10-27 16:53:11,141 | sd | INFO | sd_models | Autodetect model: detect="Stable Diffusion XL" class=StableDiffusionXLPipeline file="d:\StableDiffusion\models\pony\realistic\vanillaPonyMix_v10.safetensors" size=6617MB
2024-10-27 16:53:11,142 | sd | DEBUG | sd_models | Diffusers accelerate: direct=True
2024-10-27 16:53:14,115 | sd | DEBUG | sd_models | Setting model: pipeline=StableDiffusionXLPipeline config={'low_cpu_mem_usage': True, 'torch_dtype': torch.bfloat16, 'load_connected_pipeline': True, 'extract_ema': False, 'config': 'configs/sdxl', 'use_safetensors': True, 'cache_dir': 'C:\\Users\\Seafan\\.cache\\huggingface\\hub'}
2024-10-27 16:53:14,115 | sd | INFO | textual_inversion | Load network: type=embeddings loaded=0 skipped=0 time=0.00
2024-10-27 16:53:14,116 | sd | DEBUG | sd_models | Setting model: component=VAE slicing=True
2024-10-27 16:53:14,116 | sd | DEBUG | sd_models | Setting model: component=VAE tiling=True
2024-10-27 16:53:14,116 | sd | DEBUG | sd_models | Setting model: attention="Scaled-Dot-Product"
2024-10-27 16:53:14,130 | sd | DEBUG | sd_models | Setting model: offload=none
2024-10-27 16:53:14,130 | sd | DEBUG | shared | Read: file="E:\sdnext\configs\sdxl\vae\config.json" json=15 bytes=674 time=0.000
2024-10-27 16:53:15,114 | sd | DEBUG | devices | GC: utilization={'gpu': 35, 'ram': 20, 'threshold': 80} gc={'collected': 450, 'saved': 0.0} before={'gpu': 8.42, 'ram': 12.74} after={'gpu': 8.42, 'ram': 12.74, 'retries': 0, 'oom': 0} device=cuda fn=reload_model_weights:load_diffuser time=0.18
2024-10-27 16:53:15,120 | sd | INFO | sd_models | Load model: time=3.79 load=2.97 move=0.80 native=1024 memory={'ram': {'used': 12.74, 'total': 63.75}, 'gpu': {'used': 8.42, 'total': 23.99}, 'retries': 0, 'oom': 0}
2024-10-27 16:53:15,122 | sd | DEBUG | xyz_grid_shared | XYZ grid apply checkpoint: "pony\realistic\vanillaPonyMix_v10 [d574e497c9]"
2024-10-27 16:53:15,122 | sd | DEBUG | processing | Override: {'sd_model_checkpoint': 'pony\\realistic\\vanillaPonyMix_v10'}
2024-10-27 16:53:15,122 | sd | INFO | sd_models | Load model: select="pony\realistic\vanillaPonyMix_v10 [d574e497c9]"
2024-10-27 16:53:15,144 | sd | INFO | processing_diffusers | Base: class=StableDiffusionXLPipeline
2024-10-27 16:53:15,160 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:53:15,297 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393359]
2024-10-27 16:53:15,297 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=1/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 4, 'num_inference_steps': 30, 'eta': 1.0, 'guidance_rescale': 0.7, 'denoising_end': None, 'output_type': 'latent', 'width': 512, 'height': 512, 'parser': 'Full parser'}
2024-10-27 16:53:19,196 | sd | DEBUG | processing_class | Init hires: upscaler="Aura SR 4x" sampler="DPM SDE" resize=0x0 upscale=1126x1126
2024-10-27 16:53:19,196 | sd | INFO | processing_diffusers | Upscale: mode=1 upscaler="Aura SR 4x" context="Add with forward" resize=0x0 upscale=1126x1126
2024-10-27 16:53:19,206 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 64, 64]) dtype=torch.bfloat16 device=cuda:0 time=0.01
2024-10-27 16:53:21,699 | sd | DEBUG | images_resize | Image resize: input=<PIL.Image.Image image mode=RGB size=512x512 at 0x2B6C6199120> width=1126 height=1126 mode="Fixed" upscaler="Aura SR 4x" context="Add with forward" type=image result=<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B6C619B460> time=2.42 fn=process_hires:resize_hires
2024-10-27 16:53:21,700 | sd | DEBUG | sd_models | Pipeline class change: original=StableDiffusionXLPipeline target=StableDiffusionXLImg2ImgPipeline device=cuda:0 fn=process_diffusers:process_hires
2024-10-27 16:53:21,700 | sd | INFO | processing_diffusers | HiRes: class=StableDiffusionXLImg2ImgPipeline sampler="DPM SDE"
2024-10-27 16:53:21,718 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:53:21,874 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393359]
2024-10-27 16:53:21,875 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLImg2ImgPipeline task=DiffusersTaskType.IMAGE_2_IMAGE batch=1/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 6, 'num_inference_steps': 51, 'eta': 1.0, 'guidance_rescale': 0.7, 'output_type': 'latent', 'image': [<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B6C619B460>], 'strength': 0.3, 'parser': 'Full parser'}
2024-10-27 16:53:26,925 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 140, 140]) dtype=torch.bfloat16 device=cuda:0 time=0.326
2024-10-27 16:53:27,218 | sd | DEBUG | devices | GC: utilization={'gpu': 86, 'ram': 8, 'threshold': 80} gc={'collected': 3085, 'saved': 12.15} before={'gpu': 20.67, 'ram': 5.09} after={'gpu': 8.52, 'ram': 5.05, 'retries': 0, 'oom': 0} device=cuda fn=process_decode:vae_decode time=0.24
2024-10-27 16:53:27,241 | sd | INFO | images | Save: image="outputs\text\51722-vanillaPonyMix_v10_DPM SDE_1986393359_4_30_score 9 score 8 up score 7 up.jpg" type=JPEG width=1120 height=1120 size=215141
2024-10-27 16:53:27,254 | sd | INFO | processing_diffusers | Base: class=StableDiffusionXLPipeline
2024-10-27 16:53:27,271 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:53:27,417 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393360]
2024-10-27 16:53:27,417 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=2/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 4, 'num_inference_steps': 30, 'eta': 1.0, 'guidance_rescale': 0.7, 'denoising_end': None, 'output_type': 'latent', 'width': 512, 'height': 512, 'parser': 'Full parser'}
2024-10-27 16:53:31,367 | sd | DEBUG | processing_class | Init hires: upscaler="Aura SR 4x" sampler="DPM SDE" resize=0x0 upscale=1126x1126
2024-10-27 16:53:31,367 | sd | INFO | processing_diffusers | Upscale: mode=1 upscaler="Aura SR 4x" context="Add with forward" resize=0x0 upscale=1126x1126
2024-10-27 16:53:31,376 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 64, 64]) dtype=torch.bfloat16 device=cuda:0 time=0.008
2024-10-27 16:53:33,574 | sd | DEBUG | images_resize | Image resize: input=<PIL.Image.Image image mode=RGB size=512x512 at 0x2B6C65BCCA0> width=1126 height=1126 mode="Fixed" upscaler="Aura SR 4x" context="Add with forward" type=image result=<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B6C61990F0> time=2.17 fn=process_hires:resize_hires
2024-10-27 16:53:33,575 | sd | DEBUG | sd_models | Pipeline class change: original=StableDiffusionXLPipeline target=StableDiffusionXLImg2ImgPipeline device=cuda:0 fn=process_diffusers:process_hires
2024-10-27 16:53:33,575 | sd | INFO | processing_diffusers | HiRes: class=StableDiffusionXLImg2ImgPipeline sampler="DPM SDE"
2024-10-27 16:53:33,593 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:53:33,730 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393360]
2024-10-27 16:53:33,730 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLImg2ImgPipeline task=DiffusersTaskType.IMAGE_2_IMAGE batch=2/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 6, 'num_inference_steps': 51, 'eta': 1.0, 'guidance_rescale': 0.7, 'output_type': 'latent', 'image': [<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B6C61990F0>], 'strength': 0.3, 'parser': 'Full parser'}
2024-10-27 16:53:38,976 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 140, 140]) dtype=torch.bfloat16 device=cuda:0 time=0.326
2024-10-27 16:53:39,003 | sd | INFO | images | Save: image="outputs\text\51723-vanillaPonyMix_v10_DPM SDE_1986393360_4_30_score 9 score 8 up score 7 up.jpg" type=JPEG width=1120 height=1120 size=192359
2024-10-27 16:53:39,016 | sd | INFO | processing_diffusers | Base: class=StableDiffusionXLPipeline
2024-10-27 16:53:39,032 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:53:39,172 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393361]
2024-10-27 16:53:39,172 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=3/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 4, 'num_inference_steps': 30, 'eta': 1.0, 'guidance_rescale': 0.7, 'denoising_end': None, 'output_type': 'latent', 'width': 512, 'height': 512, 'parser': 'Full parser'}
2024-10-27 16:53:42,974 | sd | DEBUG | processing_class | Init hires: upscaler="Aura SR 4x" sampler="DPM SDE" resize=0x0 upscale=1126x1126
2024-10-27 16:53:42,974 | sd | INFO | processing_diffusers | Upscale: mode=1 upscaler="Aura SR 4x" context="Add with forward" resize=0x0 upscale=1126x1126
2024-10-27 16:53:42,982 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 64, 64]) dtype=torch.bfloat16 device=cuda:0 time=0.007
2024-10-27 16:53:45,168 | sd | DEBUG | images_resize | Image resize: input=<PIL.Image.Image image mode=RGB size=512x512 at 0x2B6387D7C40> width=1126 height=1126 mode="Fixed" upscaler="Aura SR 4x" context="Add with forward" type=image result=<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B63B43E1D0> time=2.15 fn=process_hires:resize_hires
2024-10-27 16:53:45,169 | sd | DEBUG | sd_models | Pipeline class change: original=StableDiffusionXLPipeline target=StableDiffusionXLImg2ImgPipeline device=cuda:0 fn=process_diffusers:process_hires
2024-10-27 16:53:45,169 | sd | INFO | processing_diffusers | HiRes: class=StableDiffusionXLImg2ImgPipeline sampler="DPM SDE"
2024-10-27 16:53:45,187 | sd | DEBUG | sd_samplers | Sampler: sampler="DPM SDE" class="DPMSolverSDEScheduler config={'num_train_timesteps': 1000, 'beta_start': 0.00085, 'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon', 'use_karras_sigmas': False, 'noise_sampler_seed': None, 'timestep_spacing': 'leading', 'steps_offset': 1}
2024-10-27 16:53:45,382 | sd | DEBUG | processing_helpers | Torch generator: device=cuda seeds=[1986393361]
2024-10-27 16:53:45,382 | sd | DEBUG | processing_args | Diffuser pipeline: StableDiffusionXLImg2ImgPipeline task=DiffusersTaskType.IMAGE_2_IMAGE batch=3/3x1 set={'prompt_embeds': torch.Size([1, 154, 2048]), 'pooled_prompt_embeds': torch.Size([1, 1280]), 'negative_prompt_embeds': torch.Size([1, 154, 2048]), 'negative_pooled_prompt_embeds': torch.Size([1, 1280]), 'guidance_scale': 6, 'num_inference_steps': 51, 'eta': 1.0, 'guidance_rescale': 0.7, 'output_type': 'latent', 'image': [<PIL.Image.Image image mode=RGB size=1126x1126 at 0x2B63B43E1D0>], 'strength': 0.3, 'parser': 'Full parser'}
2024-10-27 16:53:50,447 | sd | DEBUG | processing_vae | VAE decode: vae name="default" dtype=torch.bfloat16 device=cuda:0 upcast=False slicing=True tiling=True latents shape=torch.Size([1, 4, 140, 140]) dtype=torch.bfloat16 device=cuda:0 time=0.326
2024-10-27 16:53:50,679 | sd | DEBUG | devices | GC: utilization={'gpu': 86, 'ram': 8, 'threshold': 80} gc={'collected': 5526, 'saved': 12.08} before={'gpu': 20.68, 'ram': 5.1} after={'gpu': 8.6, 'ram': 5.1, 'retries': 0, 'oom': 0} device=cuda fn=process_decode:vae_decode time=0.23
2024-10-27 16:53:50,703 | sd | INFO | images | Save: image="outputs\text\51724-vanillaPonyMix_v10_DPM SDE_1986393361_4_30_score 9 score 8 up score 7 up.jpg" type=JPEG width=1120 height=1120 size=172798
2024-10-27 16:53:50,753 | sd | INFO | images | Save: image="outputs\grids\18394-vanillaPonyMix_v10_DPM SDE_1986393359_4_30_score 9 score 8 up score 7 up-grid.jpg" type=JPEG width=3360 height=1120 size=575410
2024-10-27 16:53:50,762 | sd | INFO | processing | Processed: images=4 its=3.37 time=35.58 timers={'gc': 0.47, 'init': 0.02, 'prepare': 0.07, 'encode': 7.89, 'args': 0.43, 'callback': 4.72, 'pipeline': 11.6, 'hires': 21.85, 'decode': 1.51, 'post': 0.13} memory={'ram': {'used': 5.11, 'total': 63.75}, 'gpu': {'used': 8.6, 'total': 23.99}, 'retries': 0, 'oom': 0}
2024-10-27 16:53:50,839 | sd | INFO | xyz_grid_draw | XYZ grid complete: images=3 size=(5880, 3360) time=145.45 save=0.08
2024-10-27 16:53:50,840 | sd | INFO | sd_models | Load model: select="pony\anime\tPonynai3_v6 [2b493af7c1]"
2024-10-27 16:53:52,632 | sd | DEBUG | devices | GC: utilization={'gpu': 36, 'ram': 18, 'threshold': 80} gc={'collected': 1345, 'saved': 6.92} before={'gpu': 8.6, 'ram': 11.63} after={'gpu': 1.68, 'ram': 11.63, 'retries': 0, 'oom': 0} device=cuda fn=reload_model_weights:unload_model_weights time=0.23
2024-10-27 16:53:52,632 | sd | DEBUG | sd_models | Unload weights model: {'ram': {'used': 11.63, 'total': 63.75}, 'gpu': {'used': 1.68, 'total': 23.99}, 'retries': 0, 'oom': 0}
2024-10-27 16:53:52,633 | sd | INFO | sd_models | Autodetect model: detect="Stable Diffusion XL" class=StableDiffusionXLPipeline file="d:\StableDiffusion\models\pony\anime\tPonynai3_v6.safetensors" size=6617MB
2024-10-27 16:53:52,634 | sd | DEBUG | sd_models | Diffusers accelerate: direct=True
2024-10-27 16:53:55,638 | sd | DEBUG | sd_models | Setting model: pipeline=StableDiffusionXLPipeline config={'low_cpu_mem_usage': True, 'torch_dtype': torch.bfloat16, 'load_connected_pipeline': True, 'extract_ema': False, 'config': 'configs/sdxl', 'use_safetensors': True, 'cache_dir': 'C:\\Users\\Seafan\\.cache\\huggingface\\hub'}
2024-10-27 16:53:55,639 | sd | INFO | textual_inversion | Load network: type=embeddings loaded=0 skipped=0 time=0.00
2024-10-27 16:53:55,640 | sd | DEBUG | sd_models | Setting model: component=VAE slicing=True
2024-10-27 16:53:55,640 | sd | DEBUG | sd_models | Setting model: component=VAE tiling=True
2024-10-27 16:53:55,640 | sd | DEBUG | sd_models | Setting model: attention="Scaled-Dot-Product"
2024-10-27 16:53:55,654 | sd | DEBUG | sd_models | Setting model: offload=none
2024-10-27 16:53:55,654 | sd | DEBUG | shared | Read: file="E:\sdnext\configs\sdxl\vae\config.json" json=15 bytes=674 time=0.000
2024-10-27 16:53:56,642 | sd | DEBUG | devices | GC: utilization={'gpu': 35, 'ram': 20, 'threshold': 80} gc={'collected': 448, 'saved': 0.0} before={'gpu': 8.42, 'ram': 12.84} after={'gpu': 8.42, 'ram': 12.84, 'retries': 0, 'oom': 0} device=cuda fn=reload_model_weights:load_diffuser time=0.18
2024-10-27 16:53:56,647 | sd | INFO | sd_models | Load model: time=3.82 load=3.01 move=0.80 native=1024 memory={'ram': {'used': 12.84, 'total': 63.75}, 'gpu': {'used': 8.42, 'total': 23.99}, 'retries': 0, 'oom': 0}
2024-10-27 16:53:56,732 | sd | INFO | images | Save: image="outputs\grids\grid-00171-tPonynai3_v6_DPM SDE_1986393359_4_30_score 9 score 8 up score 7 up.jpg" type=JPEG width=5880 height=3360 size=1977150
2024-10-27 16:53:56,902 | sd | INFO | processing | Processed: images=3 its=566.04 time=0.16 timers={'gc': 0.89, 'init': 6.02, 'prepare': 0.07, 'encode': 7.89, 'args': 0.43, 'callback': 4.72, 'pipeline': 11.6, 'hires': 21.85, 'decode': 1.51, 'post': 0.27} memory={'ram': {'used': 12.95, 'total': 63.75}, 'gpu': {'used': 8.42, 'total': 23.99}, 'retries': 0, 'oom': 0}
2024-10-27 16:53:59,705 | sd | DEBUG | launch | Server: alive=True jobs=1 requests=315 uptime=200 memory=12.95/63.75 backend=Backend.DIFFUSERS state=idle

Backend

Diffusers

UI

Standard

Branch

Master

Model

StableDiffusion XL

Acknowledgements

[X] I have read the above and searched for existing issues
[X] I confirm that this is classified correctly and its not an extension issue

vladmandic / automatic