Flux as refiner? #3445

SAC020 · 2024-09-19T15:07:29Z

SAC020
Sep 19, 2024

Not opening this as a defect for now but as a question

I have selected Flux1d as a refiner (native, not a flavor, tried nf4 too but same result), but it doesn't seem to do anything. This is a relevant portion of the generation log. On dev branch, updated a couple of days ago, can provide additional info.

Am I doing / setting / selecting something wrong?

18:01:52-643986 INFO     Base: class=StableDiffusionXLPipeline
18:01:52-876610 DEBUG    Sampler: sampler="DPM SDE" config={'num_train_timesteps': 1000, 'beta_start': 0.00085,
                         'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon',
                         'use_karras_sigmas': True, 'noise_sampler_seed': None, 'timestep_spacing': 'leading',
                         'steps_offset': 1}
18:01:52-967875 DEBUG    Torch generator: device=cuda seeds=[1393970976]
18:01:53-061633 DEBUG    Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=1/1x1
                         set={'prompt': 1, 'negative_prompt': 1, 'guidance_scale': 3, 'num_inference_steps': 30, 'eta':
                         1.0, 'guidance_rescale': 0.7, 'denoising_end': None, 'output_type': 'latent', 'width': 1024,
                         'height': 1024, 'ip_adapter_image': [[<PIL.PngImagePlugin.PngImageFile image mode=RGB
                         size=957x957 at 0x26E2F321350>]], 'parser': 'Fixed attention'}
18:02:00-020130 DEBUG    Server: alive=True jobs=1 requests=34820 uptime=111742 memory=14.5/63.92
                         backend=Backend.DIFFUSERS state=idle
Progress  1.57s/it █████████████████████████████████ 100% 30/30 00:47 00:00 Base
18:03:06-723277 DEBUG    Init hires: upscaler="ESRGAN 4xNMKDSuperscale_4xNMKDSuperscale" sampler="DPM SDE" resize=0x0
                         upscale=1024x1024
18:03:06-732204 INFO     Upscale: mode=1 upscaler="ESRGAN 4xNMKDSuperscale_4xNMKDSuperscale" context="Add with forward"
                         resize=0x0 upscale=1024x1024
18:03:09-704777 DEBUG    Image resize: input=<PIL.Image.Image image mode=RGB size=1024x1024 at 0x26E415E2210> width=1024
                         height=1024 mode="Fixed" upscaler="ESRGAN 4xNMKDSuperscale_4xNMKDSuperscale" context="Add with
                         forward" type=image result=<PIL.Image.Image image mode=RGB size=1024x1024 at 0x26E415E3390>
                         time=0.01 fn=C:\ai\automatic\modules\processing_helpers.py:resize_hires
18:03:09-712217 DEBUG    Pipeline class change: original=StableDiffusionXLPipeline
                         target=StableDiffusionXLImg2ImgPipeline device=cpu fn=process_diffusers
18:03:09-714698 INFO     HiRes: class=StableDiffusionXLImg2ImgPipeline sampler="DPM SDE"
18:03:09-834730 DEBUG    Sampler: sampler="DPM SDE" config={'num_train_timesteps': 1000, 'beta_start': 0.00085,
                         'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon',
                         'use_karras_sigmas': True, 'noise_sampler_seed': None, 'timestep_spacing': 'leading',
                         'steps_offset': 1}
18:03:09-839193 DEBUG    Torch generator: device=cuda seeds=[1393970976]
18:03:09-840681 DEBUG    Diffuser pipeline: StableDiffusionXLImg2ImgPipeline task=DiffusersTaskType.IMAGE_2_IMAGE
                         batch=1/1x1 set={'prompt': 1, 'negative_prompt': 1, 'guidance_scale': 6, 'num_inference_steps':
                         100, 'eta': 1.0, 'guidance_rescale': 0.7, 'output_type': 'latent', 'image': [<PIL.Image.Image
                         image mode=RGB size=1024x1024 at 0x26E415E3390>], 'strength': 0.2, 'ip_adapter_image':
                         [[<PIL.PngImagePlugin.PngImageFile image mode=RGB size=957x957 at 0x26E2F321350>]], 'width':
                         1024, 'height': 1024, 'parser': 'Fixed attention'}
Progress  1.57it/s ████████████████████████████████ 100% 20/20 00:12 00:00 Hires
18:03:29-462754 INFO     Processed: images=1 time=150.24 its=0.20 memory={'ram': {'used': 19.87, 'total': 63.92}, 'gpu':
                         {'used': 8.96, 'total': 23.99}, 'retries': 0, 'oom': 0}

vladmandic · 2024-09-19T15:50:51Z

vladmandic
Sep 19, 2024
Maintainer

this is not specific to flux, your refine step is not even being triggered, most likely your refiner start is 0, set it to a value
and to simplify until you get it working, you may want to skip resize/upscale/hires. run base + refine only.

1 reply

SAC020 Sep 19, 2024
Author

this is not specific to flux, your refine step is not even being triggered, most likely your refiner start is 0, set it to a value and to simplify until you get it working, you may want to skip resize/upscale/hires. run base + refine only.

this is how it's set. How should I set it?

vladmandic · 2024-09-19T16:19:19Z

vladmandic
Sep 19, 2024
Maintainer

exactly like i said?? disable rescale, disable hires, set refiner start to be some non-zero value

then later you can re-add upscale and hires if you really want.

1 reply

SAC020 Sep 19, 2024
Author

exactly like i said?? disable rescale, disable hires, set refiner start to be some non-zero value

then later you can re-add upscale and hires if you really want.

Ok. Identical setup.

19:48:17-673831 INFO     Select: refiner="Diffusers\sayakpaul/flux.1-dev-nf4 [b054bc66ae]"
19:48:17-678791 DEBUG    Load model: existing=False
                         target=models\Diffusers\models--sayakpaul--flux.1-dev-nf4\snapshots\b054bc66ae1097b811848c3739e
                         cd673a864bda1 info=None
19:48:18-269050 DEBUG    GC: utilization={'gpu': 40, 'ram': 12, 'threshold': 80} gc={'collected': 2166, 'saved': 1.0}
                         before={'gpu': 9.53, 'ram': 7.86} after={'gpu': 8.53, 'ram': 7.86, 'retries': 0, 'oom': 0}
                         device=cuda fn=unload_model_weights time=0.36
19:48:18-271034 DEBUG    Unload weights refiner: {'ram': {'used': 7.86, 'total': 63.92}, 'gpu': {'used': 8.53, 'total':
                         23.99}, 'retries': 0, 'oom': 0}
19:48:18-281946 DEBUG    Diffusers loading:
                         path="models\Diffusers\models--sayakpaul--flux.1-dev-nf4\snapshots\b054bc66ae1097b811848c3739ec
                         d673a864bda1"
19:48:18-282939 INFO     Autodetect: refiner="FLUX" class=FluxPipeline
                         file="models\Diffusers\models--sayakpaul--flux.1-dev-nf4\snapshots\b054bc66ae1097b811848c3739ec
                         d673a864bda1" size=0MB
19:48:18-284426 DEBUG    Loading FLUX: model="Diffusers\sayakpaul/flux.1-dev-nf4" repo="sayakpaul/flux.1-dev-nf4"
                         unet="None" t5="None" vae="None" quant=nf4 offload=model dtype=torch.float16
19:48:18-285914 DEBUG    HF login: no token provided
19:48:30-179245 DEBUG    GC: utilization={'gpu': 60, 'ram': 22, 'threshold': 80} gc={'collected': 48301, 'saved': 0.0}
                         before={'gpu': 14.37, 'ram': 13.77} after={'gpu': 14.37, 'ram': 13.77, 'retries': 0, 'oom': 0}
                         device=cuda fn=load_flux_nf4 time=0.39
19:48:30-709485 DEBUG    Loading FLUX: preloaded=['transformer']
Diffusers  8.81s/it █████████████ 100% 2/2 00:17 00:00 Loading checkpoint shards
Diffusers  2.69s/it ████████ 100% 7/7 00:18 00:00 Loading pipeline components...
19:48:49-904254 INFO     Load embeddings: loaded=2 skipped=11 time=0.02
19:48:49-994525 DEBUG    Setting refiner VAE: no-half=True
19:48:49-995516 DEBUG    Setting refiner: slicing=True
19:48:49-996509 DEBUG    Setting refiner: tiling=True
19:48:49-997500 DEBUG    Setting model: attention=Scaled-Dot-Product
19:48:50-013869 DEBUG    Setting refiner: offload=model
19:48:52-835661 DEBUG    GC: utilization={'gpu': 36, 'ram': 39, 'threshold': 80} gc={'collected': 231, 'saved': 0.0}
                         before={'gpu': 8.53, 'ram': 24.84} after={'gpu': 8.53, 'ram': 24.84, 'retries': 0, 'oom': 0}
                         device=cuda fn=load_diffuser time=0.31
19:48:52-846574 INFO     Load refiner: time=34.24 load=31.60 options=0.11 move=2.51 native=1024 {'ram': {'used': 24.84,
                         'total': 63.92}, 'gpu': {'used': 8.53, 'total': 23.99}, 'retries': 0, 'oom': 0}
19:48:52-850044 DEBUG    Setting changed: sd_model_refiner=Diffusers\sayakpaul/flux.1-dev-nf4 [b054bc66ae] progress=True
19:48:52-851533 DEBUG    Save: file="config.json" json=37 bytes=1535 time=0.001
19:48:52-855006 DEBUG    Unused settings: ['cross_attention_options']
19:49:14-441170 INFO     IP adapter: ['Plus ViT-H SDXL:0.5:0:1'] image=[[<PIL.JpegImagePlugin.JpegImageFile image
                         mode=RGB size=1024x1024 at 0x26E4156BBD0>]] mask=True time=1.99
Loading model: C:\ai\automatic\models\Lora\cydohd-sdxl-v33-civitai-epic8-prodigy-large.safetensors ━━━━━━ 0.0/2… -:--:--
                                                                                                          MB
19:49:16-961845 DEBUG    LoRA file="C:\ai\automatic\models\Lora\cydohd-sdxl-v33-civitai-epic8-prodigy-large.safetensors"
                         type={'ModuleTypeLora'} keys=986
Loading model: C:\ai\automatic\models\Lora\feet_style_v3.safetensors ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/456.5 MB -:--:--
19:49:21-065810 DEBUG    LoRA file="C:\ai\automatic\models\Lora\feet_style_v3.safetensors" type={'ModuleTypeLora'}
                         keys=986
Loading model: C:\ai\automatic\models\Lora\perfection_style.safetensors ━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/912.6 MB -:--:--
19:49:27-625051 DEBUG    LoRA file="C:\ai\automatic\models\Lora\perfection_style.safetensors" type={'ModuleTypeLora'}
                         keys=986
Loading model: C:\ai\automatic\models\Lora\detailed.safetensors ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/912.6 MB -:--:--
19:49:33-728422 DEBUG    LoRA file="C:\ai\automatic\models\Lora\detailed.safetensors" type={'ModuleTypeLora'} keys=986
Loading model: C:\ai\automatic\models\Lora\sacbf-dalle-sdxl-v31-colab-epic8-prodigy-e200.safetensors ━━━━━ 0.0/5… -:--:…
                                                                                                           MB
19:49:34-565189 DEBUG    LoRA
                         file="C:\ai\automatic\models\Lora\sacbf-dalle-sdxl-v31-colab-epic8-prodigy-e200.safetensors"
                         type={'ModuleTypeLora'} keys=986
19:49:34-575108 INFO     LoRA apply: ['cydohd-sdxl-v33-civitai-epic8-prodigy-large', 'feet_style_v3',
                         'perfection_style', 'detailed', 'sacbf-dalle-sdxl-v31-colab-epic8-prodigy-e200'] patch=0.00
                         te=[0.7, 1.0, 0.5, 0.5, 0.2] unet=[[0.7, 0.7, 0.7], [1.0, 1.0, 1.0], [0.5, 0.5, 0.5], [0.5,
                         0.5, 0.5], [0.2, 0.2, 0.2]] dims=[None, None, None, None, None] load=20.13
19:49:34-579572 INFO     Base: class=StableDiffusionXLPipeline
19:49:34-633140 DEBUG    Sampler: sampler="DPM SDE" config={'num_train_timesteps': 1000, 'beta_start': 0.00085,
                         'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon',
                         'use_karras_sigmas': True, 'noise_sampler_seed': None, 'timestep_spacing': 'leading',
                         'steps_offset': 1}
19:49:34-635620 DEBUG    Torch generator: device=cuda seeds=[2283204772]
19:49:34-637109 DEBUG    Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=1/1x1
                         set={'prompt': 1, 'negative_prompt': 1, 'guidance_scale': 3, 'num_inference_steps': 43, 'eta':
                         1.0, 'guidance_rescale': 0.7, 'denoising_end': 0.7, 'output_type': 'latent', 'width': 1024,
                         'height': 1024, 'ip_adapter_image': [[<PIL.JpegImagePlugin.JpegImageFile image mode=RGB
                         size=1024x1024 at 0x26E4156BBD0>]], 'parser': 'Fixed attention'}
Progress  1.89it/s ████████████████▊                  51% 25/49 00:13 00:12 Base
19:49:49-088774 DEBUG    Init hires: upscaler="None" sampler="DPM SDE" resize=0x0 upscale=1024x1024
19:49:49-140358 DEBUG    Pipeline class change: original=FluxPipeline target=FluxImg2ImgPipeline device=cpu
                         fn=process_diffusers
19:49:53-819206 INFO     Refiner: class=FluxImg2ImgPipeline
19:49:53-872773 WARNING  FLUX: sampler="DPM SDE" unsupported
19:49:53-874757 DEBUG    Torch generator: device=cuda seeds=[2283204772]
19:49:53-876244 DEBUG    Diffuser pipeline: FluxImg2ImgPipeline task=DiffusersTaskType.IMAGE_2_IMAGE batch=1/1x1
                         set={'prompt': 1, 'guidance_scale': 6, 'num_inference_steps': 13, 'image': [<PIL.Image.Image
                         image mode=RGB size=1024x1024 at 0x26E2F929510>], 'output_type': 'np', 'width': 1024, 'height':
                         1024, 'parser': 'Fixed attention'}
19:49:57-286291 ERROR    Exception: Input type (struct c10::Half) and bias type (float) should be the same
19:49:57-287779 ERROR    Arguments: args=('task(2yftvlsdh216iqg)', 'full body view, natural skin texture, detailed,
                         perfection, nplhd1,
                         plhd2,\n\n<lora:cydohd-sdxl-v33-civitai-epic8-prodigy-large:0.7> <lora:feet_style_v3:1.0>
                         <lora:perfection_style:0.5> <lora:detailed:0.5>
                         <lora:sacbf-dalle-sdxl-v31-colab-epic8-prodigy-e200:0.2>\n\n', '', [], 30, 13, 28, True, False,
                         False, False, 1, 1, 3, 6, 0.7, 0, 0.5, 1, 1, -1.0, -1.0, 0, 0, 0, 1024, 1024, True, 0.2, 1, 1,
                         'Add with forward', 'None', False, 20, 0, 0, 10, 0.7, 'detailed realistic', '', 0, 0, 0, 0, False, 4, 0.95, False, 0.6, 1,
                         '#000000', 0, [], 0, 1, 'Plus ViT-H SDXL', 'None', 'None', 'None', 0.5, 0.5, 0.5, 0.5,
                         [<tempfile._TemporaryFileWrapper object at 0x0000026E43A4C690>], None, None, None, False,
                         False, False, False, 0, 0, 0, 0, 1, 1, 1, 1, None, None, None, None, False, '', False, 0, '',
                         [], 0, '', [], 0, '', [], False, True, False, False, False, False, 0, 'None', [], 'FaceID
                         Base', True, True, 1, 1, 1, 0.5, True, 'person', 1, 0.5, True, 'None', 16, 'None', 1, True,
                         'None', 2, True, 1, 0, True, 'none', 3, 4, 0.25, 0.25, 'THUDM/CogVideoX-2b', 'DDIM', 49, 6,
                         'balanced', True, 'None', 8, True, 1, 0, None, None, 3, 1, 1, 0.8, 8, 64, True, 0.65, True,
                         False, 1, 1, 1, True, 0.5, 600.0, 1.0, True, None, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0.5, 0.5,
                         'OpenGVLab/InternVL-14B-224px', False, 0.7, 1.2, 128, False, False, 'positive', 'comma', 0,
                         False, False, '', 'None', '', 1, '', 'None', 1, True, 10, 'None', True, 0, 'None', 2, True, 1,
                         0, 1, 'cydohd-sdxl-v33-civitai-epic8-prodigy-large,
                         cydohd-sdxl-v34-civitai-sdxl-prodigy-large-e50,
                         cydohd-sdxl-v34-civitai-sdxl-prodigy-large-e100,
                         cydohd-sdxl-v34-civitai-sdxl-prodigy-large-e200, cydohd-sdxl-v32-colab-epic8-adamw8-large-e100,
                         cydohd-sdxl-v32-colab-epic8-adamw8-large-e200, cydohd-sdxl-v31-colab-epic8-prodigy-large-e50,
                         cydohd-sdxl-v31-colab-epic8-prodigy-large-e100,
                         cydohd-sdxl-v31-colab-epic8-prodigy-large-e150', ['Base SDXL', 'Base ViT-H SDXL', 'Plus ViT-H
                         SDXL', 'Plus Face ViT-H SDXL'], 1, '<lora:sacbf-dalle-sdxl-v31-colab-epic8-prodigy-e200:0.2>,
                         <lora:sacbf-dalle-sdxl-v13-civitai-epic7-adam-style_golden_standard:0.2>,
                         <lora:sacbf-dalle-sdxl-v26-civitai-concept-epic8-adam:0.2>,
                         <lora:sacbf-mixed-sdxl-v28-colab-adam-e200:0.2>',
                         ['cydohd-sdxl-v33-civitai-epic8-prodigy-large-000045',
                         'cydohd-sdxl-v32-colab-epic8-adamw8-large-e100',
                         'cydohd-sdxl-v32-colab-epic8-adamw8-large-e200',
                         'cydohd-sdxl-v31-colab-epic8-prodigy-large-e50',
                         'cydohd-sdxl-v31-colab-epic8-prodigy-large-e100',
                         'cydohd-sdxl-v31-colab-epic8-prodigy-large-e150'], 0, '', ['Base SDXL', 'Base ViT-H SDXL',
                         'Plus ViT-H SDXL'], False, True, False, False, True, False, 0) kwargs={}
19:49:57-296211 ERROR    gradio call: RuntimeError
╭───────────────────────────────────────── Traceback (most recent call last) ──────────────────────────────────────────╮
│ C:\ai\automatic\modules\call_queue.py:31 in f                                                                        │
│                                                                                                                      │
│   30 │   │   │   try:                                                                                                │
│ ❱ 31 │   │   │   │   res = func(*args, **kwargs)                                                                     │
│   32 │   │   │   │   progress.record_results(id_task, res)                                                           │
│                                                                                                                      │
│ C:\ai\automatic\modules\txt2img.py:93 in txt2img                                                                     │
│                                                                                                                      │
│    92 │   if processed is None:                                                                                      │
│ ❱  93 │   │   processed = processing.process_images(p)                                                               │
│    94 │   processed = scripts.scripts_txt2img.after(p, processed, *args)                                             │
│                                                                                                                      │
│ C:\ai\automatic\modules\processing.py:191 in process_images                                                          │
│                                                                                                                      │
│   190 │   │   │   with context_hypertile_vae(p), context_hypertile_unet(p):                                          │
│ ❱ 191 │   │   │   │   processed = process_images_inner(p)                                                            │
│   192                                                                                                                │
│                                                                                                                      │
│ C:\ai\automatic\modules\processing.py:313 in process_images_inner                                                    │
│                                                                                                                      │
│   312 │   │   │   │   │   from modules.processing_diffusers import process_diffusers                                 │
│ ❱ 313 │   │   │   │   │   x_samples_ddim = process_diffusers(p)                                                      │
│   314 │   │   │   │   else:                                                                                          │
│                                                                                                                      │
│ C:\ai\automatic\modules\processing_diffusers.py:281 in process_diffusers                                             │
│                                                                                                                      │
│   280 │   │   │   │   │   shared.sd_refiner.register_to_config(requires_aesthetics_score = getattr(shared.sd_refiner │
│ ❱ 281 │   │   │   │   refiner_output = shared.sd_refiner(**refiner_args) # pylint: disable=not-callable              │
│   282 │   │   │   │   if isinstance(refiner_output, dict):                                                           │
│                                                                                                                      │
│                                               ... 10 frames hidden ...                                               │
│                                                                                                                      │
│ C:\ai\automatic\venv\Lib\site-packages\torch\nn\modules\module.py:1553 in _wrapped_call_impl                         │
│                                                                                                                      │
│   1552 │   │   else:                                                                                                 │
│ ❱ 1553 │   │   │   return self._call_impl(*args, **kwargs)                                                           │
│   1554                                                                                                               │
│                                                                                                                      │
│ C:\ai\automatic\venv\Lib\site-packages\torch\nn\modules\module.py:1562 in _call_impl                                 │
│                                                                                                                      │
│   1561 │   │   │   │   or _global_forward_hooks or _global_forward_pre_hooks):                                       │
│ ❱ 1562 │   │   │   return forward_call(*args, **kwargs)                                                              │
│   1563                                                                                                               │
│                                                                                                                      │
│ C:\ai\automatic\extensions-builtin\Lora\networks.py:449 in network_Conv2d_forward                                    │
│                                                                                                                      │
│   448 │   network_apply_weights(self)                                                                                │
│ ❱ 449 │   return originals.Conv2d_forward(self, input)                                                               │
│   450                                                                                                                │
│                                                                                                                      │
│ C:\ai\automatic\venv\Lib\site-packages\torch\nn\modules\conv.py:458 in forward                                       │
│                                                                                                                      │
│    457 │   def forward(self, input: Tensor) -> Tensor:                                                               │
│ ❱  458 │   │   return self._conv_forward(input, self.weight, self.bias)                                              │
│    459                                                                                                               │
│                                                                                                                      │
│ C:\ai\automatic\venv\Lib\site-packages\torch\nn\modules\conv.py:454 in _conv_forward                                 │
│                                                                                                                      │
│    453 │   │   │   │   │   │   │   _pair(0), self.dilation, self.groups)                                             │
│ ❱  454 │   │   return F.conv2d(input, weight, bias, self.stride,                                                     │
│    455 │   │   │   │   │   │   self.padding, self.dilation, self.groups)                                             │
╰──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
RuntimeError: Input type (struct c10::Half) and bias type (float) should be the same
19:50:00-338277 DEBUG    Server: alive=True jobs=1 requests=37019 uptime=118222 memory=22.06/63.92
                         backend=Backend.DIFFUSERS state=idle

vladmandic · 2024-09-19T18:14:41Z

vladmandic
Sep 19, 2024
Maintainer

thats a type mismatch because of quantized model and incompatible offloading.
in general, when troubleshooting you need to simplify to bare minimum and then build your workflow on top of that.
you have ipadapters and loras and whatnot. first make sure your flux is working as it should on its own. before trying to do this all this.

1 reply

SAC020 Sep 22, 2024
Author

thats a type mismatch because of quantized model and incompatible offloading. in general, when troubleshooting you need to simplify to bare minimum and then build your workflow on top of that. you have ipadapters and loras and whatnot. first make sure your flux is working as it should on its own. before trying to do this all this.

ok, no loras and ip-adapter, log below (please disregard the xyz, it was set on "nothing")

you are mentioning offloading, I've also tried without --medvram, same result.

what settings should I try?

PS C:\ai\automatic> .\webui.bat --debug --medvram
Using VENV: C:\ai\automatic\venv
09:56:41-125834 INFO     Starting SD.Next
09:56:41-128811 INFO     Logger: file="C:\ai\automatic\sdnext.log" level=DEBUG size=65 mode=create
09:56:41-130298 INFO     Python version=3.11.9 platform=Windows bin="C:\ai\automatic\venv\Scripts\python.exe"
                         venv="C:\ai\automatic\venv"
09:56:41-413901 INFO     Version: app=sd.next updated=2024-09-17 hash=fedb35e9 branch=dev
                         url=https://github.com/vladmandic/automatic/tree/dev ui=dev
09:56:42-452768 INFO     Latest published version: e7ec07f9783701629ca1411ad82aec87232501b9 2024-09-13T16:51:56Z
09:56:42-466656 INFO     Platform: arch=AMD64 cpu=Intel64 Family 6 Model 165 Stepping 5, GenuineIntel system=Windows
                         release=Windows-10-10.0.22631-SP0 python=3.11.9
09:56:42-468144 DEBUG    Setting environment tuning
09:56:42-469136 INFO     HF cache folder: C:\Users\sebas\.cache\huggingface\hub
09:56:42-470128 DEBUG    Torch allocator: "garbage_collection_threshold:0.80,max_split_size_mb:512"
09:56:42-495424 DEBUG    Torch overrides: cuda=False rocm=False ipex=False diml=False openvino=False
09:56:42-497905 DEBUG    Torch allowed: cuda=True rocm=True ipex=True diml=True openvino=True
09:56:42-519729 INFO     nVidia CUDA toolkit detected: nvidia-smi present
09:56:42-751359 WARNING  Modified files: ['models/Reference/playgroundai--playground-v2-1024px-aesthetic.jpg']
09:56:42-847087 INFO     Verifying requirements
09:56:42-852048 INFO     Verifying packages
09:56:42-900160 DEBUG    Repository update time: Tue Sep 17 17:37:27 2024
09:56:42-901152 INFO     Startup: standard
09:56:42-902144 INFO     Verifying submodules
09:56:45-789152 DEBUG    Git detached head detected: folder="extensions-builtin/sd-extension-chainner" reattach=main
09:56:45-790641 DEBUG    Submodule: extensions-builtin/sd-extension-chainner / main
09:56:45-951667 DEBUG    Git detached head detected: folder="extensions-builtin/sd-extension-system-info" reattach=main
09:56:45-953154 DEBUG    Submodule: extensions-builtin/sd-extension-system-info / main
09:56:46-095507 DEBUG    Git detached head detected: folder="extensions-builtin/sd-webui-agent-scheduler" reattach=main
09:56:46-096995 DEBUG    Submodule: extensions-builtin/sd-webui-agent-scheduler / main
09:56:46-286451 DEBUG    Git detached head detected: folder="extensions-builtin/sdnext-modernui" reattach=dev
09:56:46-287443 DEBUG    Submodule: extensions-builtin/sdnext-modernui / dev
09:56:46-456539 DEBUG    Git detached head detected: folder="extensions-builtin/stable-diffusion-webui-rembg"
                         reattach=master
09:56:46-457530 DEBUG    Submodule: extensions-builtin/stable-diffusion-webui-rembg / master
09:56:46-598988 DEBUG    Git detached head detected: folder="modules/k-diffusion" reattach=master
09:56:46-600478 DEBUG    Submodule: modules/k-diffusion / master
09:56:46-736183 DEBUG    Git detached head detected: folder="wiki" reattach=master
09:56:46-737174 DEBUG    Submodule: wiki / master
09:56:46-937231 DEBUG    Register paths
09:56:47-028495 DEBUG    Installed packages: 211
09:56:47-030480 DEBUG    Extensions all: ['Lora', 'sd-extension-chainner', 'sd-extension-system-info',
                         'sd-webui-agent-scheduler', 'sd-webui-controlnet', 'sdnext-modernui',
                         'stable-diffusion-webui-images-browser', 'stable-diffusion-webui-rembg']
09:56:47-214495 DEBUG    Running extension installer:
                         C:\ai\automatic\extensions-builtin\sd-extension-system-info\install.py
09:56:47-406504 DEBUG    Running extension installer:
                         C:\ai\automatic\extensions-builtin\sd-webui-agent-scheduler\install.py
09:56:50-548597 DEBUG    Running extension installer: C:\ai\automatic\extensions-builtin\sd-webui-controlnet\install.py
09:56:58-534142 DEBUG    Running extension installer:
                         C:\ai\automatic\extensions-builtin\stable-diffusion-webui-images-browser\install.py
09:57:01-124042 DEBUG    Running extension installer:
                         C:\ai\automatic\extensions-builtin\stable-diffusion-webui-rembg\install.py
09:57:10-306304 DEBUG    Extensions all: []
09:57:10-307296 INFO     Extensions enabled: ['Lora', 'sd-extension-chainner', 'sd-extension-system-info',
                         'sd-webui-agent-scheduler', 'sd-webui-controlnet', 'sdnext-modernui',
                         'stable-diffusion-webui-images-browser', 'stable-diffusion-webui-rembg']
09:57:10-308785 INFO     Verifying requirements
09:57:10-310272 DEBUG    Setup complete without errors: 1726815430
09:57:10-325151 DEBUG    Extension preload: {'extensions-builtin': 0.01, 'extensions': 0.0}
09:57:10-326639 DEBUG    Starting module: <module 'webui' from 'C:\\ai\\automatic\\webui.py'>
09:57:10-327631 INFO     Command line args: ['--debug', '--medvram'] medvram=True debug=True
09:57:10-328623 DEBUG    Env flags: []
09:57:27-314165 INFO     Load packages: {'torch': '2.4.0+cu124', 'diffusers': '0.31.0.dev0', 'gradio': '3.43.2',
                         'transformers': '4.44.2', 'accelerate': '0.34.2'}
09:57:28-663780 DEBUG    Read: file="config.json" json=37 bytes=1646 time=0.000
09:57:28-666260 DEBUG    Unknown settings: ['cross_attention_options']
09:57:28-668740 INFO     Engine: backend=Backend.DIFFUSERS compute=cuda device=cuda attention="Scaled-Dot-Product"
                         mode=no_grad
09:57:28-747109 INFO     Device: device=NVIDIA GeForce RTX 4090 n=1 arch=sm_90 capability=(8, 9) cuda=12.4 cudnn=90100
                         driver=561.09
09:57:28-751078 DEBUG    Read: file="html\reference.json" json=52 bytes=29118 time=0.002
09:57:30-109123 DEBUG    ONNX: version=1.19.2 provider=CUDAExecutionProvider, available=['AzureExecutionProvider',
                         'CPUExecutionProvider']
09:57:30-630424 DEBUG    Importing LDM
09:57:30-676373 DEBUG    Entering start sequence
09:57:30-681864 DEBUG    Initializing
09:57:30-782622 INFO     Available VAEs: path="models\VAE" items=0
09:57:30-784111 DEBUG    Available UNets: path="models\UNET" items=0
09:57:30-785599 DEBUG    Available T5s: path="models\T5" items=0
09:57:30-787087 INFO     Disabled extensions: ['sd-webui-controlnet', 'sdnext-modernui']
09:57:30-796014 DEBUG    Read: file="cache.json" json=2 bytes=10767 time=0.001
09:57:30-810895 DEBUG    Read: file="metadata.json" json=675 bytes=2414573 time=0.012
09:57:30-856526 DEBUG    Scanning diffusers cache: folder=models\Diffusers items=2 time=0.01
09:57:30-858014 INFO     Available models: path="models\Stable-diffusion" items=14 time=0.07
09:57:31-745698 DEBUG    Load extensions
09:57:31-939143 INFO     Extension: script='extensions-builtin\Lora\scripts\lora_script.py'
                         [2;36m09:57:31-932088[0m[2;36m [0m[34mINFO    [0m LoRA networks: [33mavailable[0m=[1;36m123[0m
                         [33mfolders[0m=[1;36m4[0m
09:57:32-949161 INFO     Extension: script='extensions-builtin\sd-webui-agent-scheduler\scripts\task_scheduler.py' Using
                         sqlite file: extensions-builtin\sd-webui-agent-scheduler\task_scheduler.sqlite3
09:57:33-259567 DEBUG    Extensions init time: 1.51 Lora=0.06 sd-extension-chainner=0.10 sd-webui-agent-scheduler=0.89
                         stable-diffusion-webui-images-browser=0.30
09:57:33-300215 DEBUG    Read: file="html/upscalers.json" json=4 bytes=2672 time=0.001
09:57:33-302605 DEBUG    Read: file="extensions-builtin\sd-extension-chainner\models.json" json=24 bytes=2719 time=0.001
09:57:33-308246 DEBUG    chaiNNer models: path="models\chaiNNer" defined=24 discovered=0 downloaded=8
09:57:33-310861 DEBUG    Upscaler type=ESRGAN folder="models\ESRGAN" model="1x-ITF-SkinDiffDetail-Lite-v1"
                         path="models\ESRGAN\1x-ITF-SkinDiffDetail-Lite-v1.pth"
09:57:33-312115 DEBUG    Upscaler type=ESRGAN folder="models\ESRGAN" model="4xNMKDSuperscale_4xNMKDSuperscale"
                         path="models\ESRGAN\4xNMKDSuperscale_4xNMKDSuperscale.pth"
09:57:33-313794 DEBUG    Upscaler type=ESRGAN folder="models\ESRGAN" model="4x_NMKD-Siax_200k"
                         path="models\ESRGAN\4x_NMKD-Siax_200k.pth"
09:57:33-317271 DEBUG    Load upscalers: total=56 downloaded=11 user=3 time=0.06 ['None', 'Lanczos', 'Nearest',
                         'ChaiNNer', 'AuraSR', 'ESRGAN', 'LDSR', 'RealESRGAN', 'SCUNet', 'SD', 'SwinIR']
09:57:33-338615 DEBUG    Load styles: folder="models\styles" items=288 time=0.02
09:57:33-346551 DEBUG    Creating UI
09:57:33-348039 DEBUG    UI themes available: type=Standard themes=12
09:57:33-349527 INFO     UI theme: type=Standard name="black-teal"
09:57:33-361928 DEBUG    UI theme: css="C:\ai\automatic\javascript\black-teal.css" base="sdnext.css" user="None"
09:57:33-366392 DEBUG    UI initialize: txt2img
09:57:33-449224 DEBUG    Networks: page='model' items=65 subfolders=2 tab=txt2img folders=['models\\Stable-diffusion',
                         'models\\Diffusers', 'models\\Reference'] list=0.06 thumb=0.01 desc=0.00 info=0.00 workers=4
                         sort=Default
09:57:33-463111 DEBUG    Networks: page='lora' items=123 subfolders=0 tab=txt2img folders=['models\\Lora',
                         'models\\LyCORIS'] list=0.06 thumb=0.02 desc=0.03 info=0.03 workers=4 sort=Default
09:57:33-495188 DEBUG    Networks: page='style' items=288 subfolders=1 tab=txt2img folders=['models\\styles', 'html']
                         list=0.05 thumb=0.00 desc=0.00 info=0.00 workers=4 sort=Default
09:57:33-500177 DEBUG    Networks: page='embedding' items=13 subfolders=0 tab=txt2img folders=['models\\embeddings']
                         list=0.03 thumb=0.01 desc=0.01 info=0.00 workers=4 sort=Default
09:57:33-502132 DEBUG    Networks: page='vae' items=0 subfolders=0 tab=txt2img folders=['models\\VAE'] list=0.00
                         thumb=0.00 desc=0.00 info=0.00 workers=4 sort=Default
09:57:33-601828 DEBUG    UI initialize: img2img
09:57:33-880581 DEBUG    UI initialize: control models=models\control
09:57:34-191561 DEBUG    Read: file="ui-config.json" json=0 bytes=2 time=0.000
09:57:34-302169 DEBUG    UI themes available: type=Standard themes=12
09:57:34-897929 DEBUG    Reading failed: C:\ai\automatic\html\extensions.json [Errno 2] No such file or directory:
                         'C:\\ai\\automatic\\html\\extensions.json'
09:57:34-898922 INFO     Extension list is empty: refresh required
09:57:35-520826 DEBUG    Extension list: processed=8 installed=8 enabled=6 disabled=2 visible=8 hidden=0
09:57:35-906360 DEBUG    Root paths: ['C:\\ai\\automatic']
09:57:36-017969 INFO     Local URL: http://127.0.0.1:7860/
09:57:36-018961 DEBUG    Gradio functions: registered=2435
09:57:36-022359 DEBUG    FastAPI middleware: ['Middleware', 'Middleware']
09:57:36-025336 DEBUG    Creating API
09:57:36-198948 INFO     [AgentScheduler] Task queue is empty
09:57:36-199940 INFO     [AgentScheduler] Registering APIs
09:57:36-325924 DEBUG    Scripts setup: ['IP Adapters:0.024', 'XYZ Grid:0.014', 'Face:0.013', 'AnimateDiff:0.007',
                         'CogVideoX:0.008', 'LUT Color grading:0.005', 'X/Y/Z Grid:0.012', 'Image-to-Video:0.006',
                         'Stable Video Diffusion:0.005']
09:57:36-327411 DEBUG    Model metadata: file="metadata.json" no changes
09:57:36-328404 DEBUG    Torch mode: deterministic=False
09:57:36-440958 INFO     Torch override VAE dtype: no-half set
09:57:36-442447 DEBUG    Desired Torch parameters: dtype=FP16 no-half=False no-half-vae=True upscast=False
09:57:36-443933 INFO     Setting Torch parameters: device=cuda dtype=torch.float16 vae=torch.float32 unet=torch.float16
                         context=no_grad fp16=True bf16=None optimization=Scaled-Dot-Product
09:57:36-445919 DEBUG    Model requested: fn=<lambda>
09:57:36-446909 INFO     Select: model="epicrealismXL_v8Kiss [04639d4084]"
09:57:36-448398 DEBUG    Load model: existing=False
                         target=C:\ai\automatic\models\Stable-diffusion\epicrealismXL_v8Kiss.safetensors info=None
09:57:36-449886 DEBUG    Diffusers loading:
                         path="C:\ai\automatic\models\Stable-diffusion\epicrealismXL_v8Kiss.safetensors"
09:57:36-450878 INFO     Autodetect: model="Stable Diffusion XL" class=StableDiffusionXLPipeline
                         file="C:\ai\automatic\models\Stable-diffusion\epicrealismXL_v8Kiss.safetensors" size=6617MB
Diffusers  5.73it/s ████████ 100% 7/7 00:01 00:00 Loading pipeline components...
09:57:38-326637 DEBUG    Setting model: pipeline=StableDiffusionXLPipeline config={'low_cpu_mem_usage': True,
                         'torch_dtype': torch.float16, 'load_connected_pipeline': True, 'extract_ema': False, 'config':
                         'configs/sdxl', 'use_safetensors': True, 'cache_dir':
                         'C:\\Users\\sebas\\.cache\\huggingface\\hub'}
09:57:42-036257 INFO     Load embeddings: loaded=2 skipped=11 time=3.70
09:57:42-354052 DEBUG    Setting model VAE: no-half=True
09:57:42-355541 DEBUG    Setting model: slicing=True
09:57:42-356532 DEBUG    Setting model: tiling=True
09:57:42-357028 DEBUG    Setting model: attention=Scaled-Dot-Product
09:57:42-376868 DEBUG    Setting model: offload=model
09:57:42-413572 DEBUG    Read: file="C:\ai\automatic\configs\sdxl\vae\config.json" json=15 bytes=674 time=0.000
09:57:42-648704 DEBUG    GC: utilization={'gpu': 7, 'ram': 4, 'threshold': 80} gc={'collected': 153, 'saved': 0.0}
                         before={'gpu': 1.6, 'ram': 2.48} after={'gpu': 1.6, 'ram': 2.48, 'retries': 0, 'oom': 0}
                         device=cuda fn=load_diffuser time=0.23
09:57:42-657604 INFO     Load model: time=5.97 load=1.88 embeddings=3.71 options=0.34 native=1024 {'ram': {'used': 2.48,
                         'total': 63.92}, 'gpu': {'used': 1.6, 'total': 23.99}, 'retries': 0, 'oom': 0}
09:57:42-660580 DEBUG    Script callback init time: image_browser.py:ui_tabs=0.43 system-info.py:app_started=0.07
                         task_scheduler.py:app_started=0.14
09:57:42-662068 INFO     Startup time: 32.32 torch=10.60 accelerate=0.16 gradio=3.35 diffusers=1.93 libraries=4.26
                         samplers=0.10 extensions=1.51 models=0.07 face-restore=0.89 upscalers=0.06 ui-networks=0.43
                         ui-txt2img=0.08 ui-img2img=0.07 ui-control=0.13 ui-settings=0.26 ui-extensions=1.10
                         ui-defaults=0.32 launch=0.17 api=0.09 app-started=0.21 checkpoint=6.33
09:57:42-664052 DEBUG    Save: file="config.json" json=37 bytes=1597 time=0.003
09:57:42-667028 DEBUG    Unused settings: ['cross_attention_options']
09:57:47-819538 INFO     Select: refiner="Diffusers\sayakpaul/flux.1-dev-nf4 [b054bc66ae]"
09:57:47-821026 DEBUG    Load model: existing=False
                         target=models\Diffusers\models--sayakpaul--flux.1-dev-nf4\snapshots\b054bc66ae1097b811848c3739e
                         cd673a864bda1 info=None
09:57:47-822514 DEBUG    Diffusers loading:
                         path="models\Diffusers\models--sayakpaul--flux.1-dev-nf4\snapshots\b054bc66ae1097b811848c3739ec
                         d673a864bda1"
09:57:47-823506 INFO     Autodetect: refiner="FLUX" class=FluxPipeline
                         file="models\Diffusers\models--sayakpaul--flux.1-dev-nf4\snapshots\b054bc66ae1097b811848c3739ec
                         d673a864bda1" size=0MB
09:57:47-828466 DEBUG    Loading FLUX: model="Diffusers\sayakpaul/flux.1-dev-nf4" repo="sayakpaul/flux.1-dev-nf4"
                         unet="None" t5="None" vae="None" quant=nf4 offload=model dtype=torch.float16
09:57:47-829954 DEBUG    HF login: no token provided
09:57:59-122549 DEBUG    GC: utilization={'gpu': 34, 'ram': 14, 'threshold': 80} gc={'collected': 269, 'saved': 0.0}
                         before={'gpu': 8.12, 'ram': 8.72} after={'gpu': 8.12, 'ram': 8.72, 'retries': 0, 'oom': 0}
                         device=cuda fn=load_flux_nf4 time=0.27
09:57:59-642357 DEBUG    Loading FLUX: preloaded=['transformer']
09:57:59-716275 DEBUG    Server: alive=True jobs=1 requests=21 uptime=31 memory=2.49/63.92 backend=Backend.DIFFUSERS
                         state=idle
Diffusers 10.32s/it █████████████ 100% 2/2 00:20 00:00 Loading checkpoint shards
Diffusers  3.22s/it ████████ 100% 7/7 00:22 00:00 Loading pipeline components...
09:58:23-010264 INFO     Load embeddings: loaded=2 skipped=11 time=0.04
09:58:23-246360 DEBUG    Setting refiner VAE: no-half=True
09:58:23-247847 DEBUG    Setting refiner: slicing=True
09:58:23-249335 DEBUG    Setting refiner: tiling=True
09:58:23-250328 DEBUG    Setting model: attention=Scaled-Dot-Product
09:58:23-275128 DEBUG    Setting refiner: offload=model
09:58:30-222218 DEBUG    GC: utilization={'gpu': 7, 'ram': 31, 'threshold': 80} gc={'collected': 359, 'saved': 0.0}
                         before={'gpu': 1.64, 'ram': 20.11} after={'gpu': 1.64, 'ram': 20.11, 'retries': 0, 'oom': 0}
                         device=cuda fn=load_diffuser time=0.33
09:58:30-240072 INFO     Load refiner: time=42.07 load=35.15 options=0.26 move=6.61 native=1024 {'ram': {'used': 20.11,
                         'total': 63.92}, 'gpu': {'used': 1.64, 'total': 23.99}, 'retries': 0, 'oom': 0}
09:58:30-248999 INFO     Base: class=StableDiffusionXLPipeline
09:58:30-366084 DEBUG    Sampler: sampler="DPM SDE" config={'num_train_timesteps': 1000, 'beta_start': 0.00085,
                         'beta_end': 0.012, 'beta_schedule': 'scaled_linear', 'prediction_type': 'epsilon',
                         'use_karras_sigmas': True, 'noise_sampler_seed': None, 'timestep_spacing': 'leading',
                         'steps_offset': 1}
09:58:30-397799 DEBUG    Torch generator: device=cuda seeds=[629145906]
09:58:30-399783 DEBUG    Diffuser pipeline: StableDiffusionXLPipeline task=DiffusersTaskType.TEXT_2_IMAGE batch=1/1x1
                         set={'prompt': 1, 'negative_prompt': 1, 'guidance_scale': 3, 'num_inference_steps': 43, 'eta':
                         1.0, 'guidance_rescale': 0.7, 'denoising_end': 0.7, 'output_type': 'latent', 'width': 1024,
                         'height': 1024, 'parser': 'Fixed attention'}
Progress  5.73s/it █▍                                   4% 2/49 00:13 04:29 Base09:58:48-215357 DEBUG    VAE load: type=taesd model=models\TAESD\taesdxl_decoder.pth
Progress  1.17it/s ████████████████▊                  51% 25/49 00:21 00:20 Base
09:58:58-206436 DEBUG    Init hires: upscaler="None" sampler="DPM SDE" resize=0x0 upscale=1024x1024
09:58:58-257524 DEBUG    Pipeline class change: original=FluxPipeline target=FluxImg2ImgPipeline device=cpu
                         fn=process_diffusers
09:59:00-983088 INFO     Refiner: class=FluxImg2ImgPipeline
09:59:01-039136 WARNING  FLUX: sampler="DPM SDE" unsupported
09:59:01-041616 DEBUG    Torch generator: device=cuda seeds=[629145906]
09:59:01-043104 DEBUG    Diffuser pipeline: FluxImg2ImgPipeline task=DiffusersTaskType.IMAGE_2_IMAGE batch=1/1x1
                         set={'prompt': 1, 'guidance_scale': 6, 'num_inference_steps': 13, 'image': [<PIL.Image.Image
                         image mode=RGB size=1024x1024 at 0x2892BF42ED0>], 'output_type': 'np', 'width': 1024, 'height':
                         1024, 'parser': 'Fixed attention'}
09:59:04-988360 ERROR    Exception: Input type (struct c10::Half) and bias type (float) should be the same
09:59:04-990343 ERROR    Arguments: args=('task(ijvsyqdl6460pxl)', 'full body view, natural skin texture, detailed,
                         perfection, \n\nwoman\n\non the side of a road, \n\n\n\n', '', [], 30, 13, 28, True, False,
                         False, False, 1, 1, 3, 6, 0.7, 0, 0.5, 1, 1, -1.0, -1.0, 0, 0, 0, 1024, 1024, True, 0.2, 1, 1,
                         'Add with forward', 'None', False, 20, 0, 0, 10, 0.7, '', '', 0, 0, 0, 0, False, 4, 0.95,
                         False, 0.6, 1, '#000000', 0, [], 0, 1, 'None', 'None', 'None', 'None', 0.5, 0.5, 0.5, 0.5,
                         None, None, None, None, False, False, False, False, 0, 0, 0, 0, 1, 1, 1, 1, None, None, None,
                         None, False, '', False, 0, '', [], 0, '', [], 0, '', [], False, True, False, False, False,
                         False, 0, 'None', [], 'FaceID Base', True, True, 1, 1, 1, 0.5, True, 'person', 1, 0.5, True,
                         'None', 16, 'None', 1, True, 'None', 2, True, 1, 0, True, 'none', 3, 4, 0.25, 0.25,
                         'THUDM/CogVideoX-2b', 'DDIM', 49, 6, 'balanced', True, 'None', 8, True, 1, 0, None, None, 3, 1,
                         1, 0.8, 8, 64, True, 0.65, True, False, 1, 1, 1, True, 0.5, 600.0, 1.0, True, None, 1, 0, 0, 0,
                         0, 0, 0, 0, 1, 1, 1, 0.5, 0.5, 'OpenGVLab/InternVL-14B-224px', False, 0.7, 1.2, 128, False,
                         False, 'positive', 'comma', 0, False, False, '', 'None', '', 1, '', 'None', 1, True, 10,
                         'None', True, 0, 'None', 2, True, 1, 0, 1, 'cydohd-sdxl-v33-civitai-epic8-prodigy-large,
                         cydohd-sdxl-v34-civitai-sdxl-prodigy-large-e50,
                         cydohd-sdxl-v34-civitai-sdxl-prodigy-large-e100,
                         cydohd-sdxl-v34-civitai-sdxl-prodigy-large-e200, cydohd-sdxl-v32-colab-epic8-adamw8-large-e100,
                         cydohd-sdxl-v32-colab-epic8-adamw8-large-e200, cydohd-sdxl-v31-colab-epic8-prodigy-large-e50,
                         cydohd-sdxl-v31-colab-epic8-prodigy-large-e100,
                         cydohd-sdxl-v31-colab-epic8-prodigy-large-e150', ['Base SDXL', 'Base ViT-H SDXL', 'Plus ViT-H
                         SDXL', 'Plus Face ViT-H SDXL'], 1, '<lora:sacbf-dalle-sdxl-v31-colab-epic8-prodigy-e200:0.2>,
                         <lora:sacbf-dalle-sdxl-v13-civitai-epic7-adam-style_golden_standard:0.2>,
                         <lora:sacbf-dalle-sdxl-v26-civitai-concept-epic8-adam:0.2>,
                         <lora:sacbf-mixed-sdxl-v28-colab-adam-e200:0.2>',
                         ['cydohd-sdxl-v33-civitai-epic8-prodigy-large-000045',
                         'cydohd-sdxl-v32-colab-epic8-adamw8-large-e100',
                         'cydohd-sdxl-v32-colab-epic8-adamw8-large-e200',
                         'cydohd-sdxl-v31-colab-epic8-prodigy-large-e50',
                         'cydohd-sdxl-v31-colab-epic8-prodigy-large-e100',
                         'cydohd-sdxl-v31-colab-epic8-prodigy-large-e150'], 0, '', ['Base SDXL', 'Base ViT-H SDXL',
                         'Plus ViT-H SDXL'], False, True, False, False, True, False, 0) kwargs={}
09:59:04-999271 ERROR    gradio call: RuntimeError
╭───────────────────────────────────────── Traceback (most recent call last) ──────────────────────────────────────────╮
│ C:\ai\automatic\modules\call_queue.py:31 in f                                                                        │
│                                                                                                                      │
│   30 │   │   │   try:                                                                                                │
│ ❱ 31 │   │   │   │   res = func(*args, **kwargs)                                                                     │
│   32 │   │   │   │   progress.record_results(id_task, res)                                                           │
│                                                                                                                      │
│ C:\ai\automatic\modules\txt2img.py:93 in txt2img                                                                     │
│                                                                                                                      │
│    92 │   if processed is None:                                                                                      │
│ ❱  93 │   │   processed = processing.process_images(p)                                                               │
│    94 │   processed = scripts.scripts_txt2img.after(p, processed, *args)                                             │
│                                                                                                                      │
│ C:\ai\automatic\modules\processing.py:191 in process_images                                                          │
│                                                                                                                      │
│   190 │   │   │   with context_hypertile_vae(p), context_hypertile_unet(p):                                          │
│ ❱ 191 │   │   │   │   processed = process_images_inner(p)                                                            │
│   192                                                                                                                │
│                                                                                                                      │
│ C:\ai\automatic\modules\processing.py:313 in process_images_inner                                                    │
│                                                                                                                      │
│   312 │   │   │   │   │   from modules.processing_diffusers import process_diffusers                                 │
│ ❱ 313 │   │   │   │   │   x_samples_ddim = process_diffusers(p)                                                      │
│   314 │   │   │   │   else:                                                                                          │
│                                                                                                                      │
│ C:\ai\automatic\modules\processing_diffusers.py:281 in process_diffusers                                             │
│                                                                                                                      │
│   280 │   │   │   │   │   shared.sd_refiner.register_to_config(requires_aesthetics_score = getattr(shared.sd_refiner │
│ ❱ 281 │   │   │   │   refiner_output = shared.sd_refiner(**refiner_args) # pylint: disable=not-callable              │
│   282 │   │   │   │   if isinstance(refiner_output, dict):                                                           │
│                                                                                                                      │
│                                               ... 9 frames hidden ...                                                │
│                                                                                                                      │
│ C:\ai\automatic\venv\Lib\site-packages\diffusers\models\autoencoders\vae.py:143 in forward                           │
│                                                                                                                      │
│   142 │   │                                                                                                          │
│ ❱ 143 │   │   sample = self.conv_in(sample)                                                                          │
│   144                                                                                                                │
│                                                                                                                      │
│ C:\ai\automatic\venv\Lib\site-packages\torch\nn\modules\module.py:1553 in _wrapped_call_impl                         │
│                                                                                                                      │
│   1552 │   │   else:                                                                                                 │
│ ❱ 1553 │   │   │   return self._call_impl(*args, **kwargs)                                                           │
│   1554                                                                                                               │
│                                                                                                                      │
│ C:\ai\automatic\venv\Lib\site-packages\torch\nn\modules\module.py:1562 in _call_impl                                 │
│                                                                                                                      │
│   1561 │   │   │   │   or _global_forward_hooks or _global_forward_pre_hooks):                                       │
│ ❱ 1562 │   │   │   return forward_call(*args, **kwargs)                                                              │
│   1563                                                                                                               │
│                                                                                                                      │
│ C:\ai\automatic\venv\Lib\site-packages\torch\nn\modules\conv.py:458 in forward                                       │
│                                                                                                                      │
│    457 │   def forward(self, input: Tensor) -> Tensor:                                                               │
│ ❱  458 │   │   return self._conv_forward(input, self.weight, self.bias)                                              │
│    459                                                                                                               │
│                                                                                                                      │
│ C:\ai\automatic\venv\Lib\site-packages\torch\nn\modules\conv.py:454 in _conv_forward                                 │
│                                                                                                                      │
│    453 │   │   │   │   │   │   │   _pair(0), self.dilation, self.groups)                                             │
│ ❱  454 │   │   return F.conv2d(input, weight, bias, self.stride,                                                     │
│    455 │   │   │   │   │   │   self.padding, self.dilation, self.groups)                                             │
╰──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
RuntimeError: Input type (struct c10::Half) and bias type (float) should be the same

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Flux as refiner? #3445

{{title}}

Replies: 3 comments 3 replies

{{title}}

{{title}}

{{title}}

{{title}}

{{editor}}'s edit

{{editor}}'s edit

{{title}}

{{title}}

{{editor}}'s edit

{{editor}}'s edit

Select a reply

Flux as refiner? #3445

SAC020 Sep 19, 2024

Replies: 3 comments · 3 replies

vladmandic Sep 19, 2024 Maintainer

SAC020 Sep 19, 2024 Author

vladmandic Sep 19, 2024 Maintainer

SAC020 Sep 19, 2024 Author

vladmandic Sep 19, 2024 Maintainer

SAC020 Sep 22, 2024 Author

SAC020
Sep 19, 2024

Replies: 3 comments 3 replies

vladmandic
Sep 19, 2024
Maintainer

SAC020 Sep 19, 2024
Author

vladmandic
Sep 19, 2024
Maintainer

SAC020 Sep 19, 2024
Author

vladmandic
Sep 19, 2024
Maintainer

SAC020 Sep 22, 2024
Author