Hi, I'm getting this error message. Please help. Thanks

Error occurred when executing SeargeMagicBox: 'NoneType' object has no attribute 'model' File "Y:\ComfyUI\ComfyUI\execution.py", line 152, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) File "Y:\ComfyUI\ComfyUI\execution.py", line 82, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) File "Y:\ComfyUI\ComfyUI\execution.py", line 75, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) File "Y:\ComfyUI\ComfyUI\custom_nodes\SeargeSDXL\modules\magic_box.py", line 269, in process (data, stage_result) = self.run_stage(stage, data, stage_input) File "Y:\ComfyUI\ComfyUI\custom_nodes\SeargeSDXL\modules\magic_box.py", line 248, in run_stage (data, stage_result) = stage_processor.process(data, stage_input) File "Y:\ComfyUI\ComfyUI\custom_nodes\SeargeSDXL\modules\stage_load_checkpoints.py", line 140, in process refiner_model = NodeWrapper.freeu.patch(refiner_model, b1, b2, s1, s2)[0] File "Y:\ComfyUI\ComfyUI\comfy_extras\nodes_freelunch.py", line 40, in patch model_channels = model.model.model_config.unet_config["model_channels"]

I reinstalled redownloaded twice. even tried with manager.

started with new comfyui release unzipped.
then normal git pull in node folder.

Then i get red boxes telling me none of the nodes work when i load workflow. and this in the cmd window.

Traceback (most recent call last): File "C:\comf\ComfyUI_windows_portable\ComfyUI\nodes.py", line 1725, in load_custom_node module_spec.loader.exec_module(module) File "<frozen importlib._bootstrap_external>", line 883, in exec_module File "<frozen importlib._bootstrap>", line 241, in callwith_frames_removed File "C:\comf\ComfyUI_windows_portable\ComfyUI\custom_nodes\SeargeSDXL__init.py", line 35, in <module> from .searge_sdxl import SEARGE_CLASS_MAPPINGS, SEARGE_DISPLAY_NAME_MAPPINGS File "C:\comf\ComfyUI_windows_portable\ComfyUI\custom_nodes\SeargeSDXL\searge_sdxl.py", line 32, in <module> from .modules.controlnet_adapter import SeargeControlnetAdapterV2 File "C:\comf\ComfyUI_windows_portable\ComfyUI\custom_nodes\SeargeSDXL\modules\controlnet_adapter.py", line 31, in <module> from .controlnet import canny File "C:\comf\ComfyUI_windows_portable\ComfyUI\custom_nodes\SeargeSDXL\modules\controlnet__init.py", line 4, in <module> from .canny import CannyDetector File "C:\comf\ComfyUI_windows_portable\ComfyUI\custom_nodes\SeargeSDXL\modules\controlnet\canny__init__.py", line 2, in <module> import cv2 ModuleNotFoundError: No module named 'cv2' Cannot import C:\comf\ComfyUI_windows_portable\ComfyUI\custom_nodes\SeargeSDXL module for custom nodes: No module named 'cv2'

Thanks dude... not sure if anyone asked this before, but is there a way for a batch img2img? For example: I have a folder with 1000 images and I want to do img2img on all of them (without loading manually each image)...

This is absolutely amazing! Great job - it's clear you put a lot of work and effort into this :D You are a blessing to this community!

Hi, I keep getting this error. Please help.

Error occurred when executing SeargeMagicBox: not enough values to unpack (expected 3, got 2) File "C:\Users\rkran\AI-StableDiffusion\ComfyUI_windows_portable\ComfyUI\execution.py", line 152, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) File "C:\Users\rkran\AI-StableDiffusion\ComfyUI_windows_portable\ComfyUI\execution.py", line 82, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) File "C:\Users\rkran\AI-StableDiffusion\ComfyUI_windows_portable\ComfyUI\execution.py", line 75, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) File "C:\Users\rkran\AI-StableDiffusion\ComfyUI_windows_portable\ComfyUI\custom_nodes\SeargeSDXL\modules\magic_box.py", line 269, in process (data, stage_result) = self.run_stage(stage, data, stage_input) File "C:\Users\rkran\AI-StableDiffusion\ComfyUI_windows_portable\ComfyUI\custom_nodes\SeargeSDXL\modules\magic_box.py", line 248, in run_stage (data, stage_result) = stage_processor.process(data, stage_input) File "C:\Users\rkran\AI-StableDiffusion\ComfyUI_windows_portable\ComfyUI\custom_nodes\SeargeSDXL\modules\stage_latent_inputs.py", line 127, in process (width, height) = get_mask_size(mask) File "C:\Users\rkran\AI-StableDiffusion\ComfyUI_windows_portable\ComfyUI\custom_nodes\SeargeSDXL\modules\utils.py", line 51, in get_mask_size (_, height, width) = mask.shape

I have the latest version '4.2.1'. Thanks!

This is great, thank you......

Everything works like charms,

Request to adopt also IP adapter option for other option of ReVision so we could have it integrated in this marvelous workflow.

I'm so done with this. Comfyui hasn't produced anything but garbage for a week and I'm past caring why, or why the rollbacks don't help, or any of it. I'm sick to death of this janky fucking garbage.

Error occurred when executing SeargeControlnetAdapterV2: stat: path should be string, bytes, os.PathLike or integer, not NoneType File "/media/comfydrive/comfyui3/ComfyUI/execution.py", line 152, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) File "/media/comfydrive/comfyui3/ComfyUI/execution.py", line 82, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) File "/media/comfydrive/comfyui3/ComfyUI/execution.py", line 75, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) File "/media/comfydrive/comfyui3/ComfyUI/custom_nodes/SeargeSDXL/modules/controlnet_adapter.py", line 149, in get_value (stack_entry, image) = self.create_dict( File "/media/comfydrive/comfyui3/ComfyUI/custom_nodes/SeargeSDXL/modules/controlnet_adapter.py", line 110, in create_dict cn_image = self.process_image(cn_image, controlnet_mode, low_threshold, high_threshold) File "/media/comfydrive/comfyui3/ComfyUI/custom_nodes/SeargeSDXL/modules/controlnet_adapter.py", line 84, in process_image image = leres(image, low_threshold, high_threshold, self.leres_annotator_full_path) File "/media/comfydrive/comfyui3/ComfyUI/custom_nodes/SeargeSDXL/modules/controlnet/__init__.py", line 23, in leres annotator = LeresDetector.from_pretrained(annotator_model).to(model_management.get_torch_device()) File "/media/comfydrive/comfyui3/ComfyUI/custom_nodes/SeargeSDXL/modules/controlnet/leres/__init__.py", line 23, in from_pretrained if os.path.isdir(pretrained_model_or_path): File "/usr/lib/python3.10/genericpath.py", line 42, in isdir st = os.stat(s)

I'd like to request that the user be given the option to change the sampler when running the Hires fix if possible.

The reason for this is that in my testing, whilst DPM++ SDE tends to beat Euler out on image quality when it comes to generating images on the first pass, Euler blows everything else completely out of the water when it comes to the Hires pass, so being able to use DPM++ SDE on the first pass, and then switch to Euler for the second pass, seems like it would likely boost the quality of the final image drastically.

EDIT: By modifying the hires stage node to force the usage of Euler, I can confirm that generating the first step in DPM++ SDE and then using Euler in the hires step produces better results than either Euler or DPM++ SDE for both steps

Is anyone else getting this error?
When loading the graph, the following node types were not found:

• SeargePreviewImage

• SeargeTextInputV2

• SeargeMagicBox

• SeargeDebugPrinter

• SeargePipelineTerminator

• SeargeGenerationParameters

• SeargeImageSaving

• SeargeOperatingMode

• SeargeImage2ImageAndInpainting

• SeargeModelSelector

• SeargeConditionMixing

• SeargeCustomPromptMode

• SeargeCustomAfterVaeDecode

• SeargeSeparator

• SeargeConditioningParameters

• SeargeHighResolution

• SeargeAdvancedParameters

• SeargePromptStyles

• SeargeCustomAfterUpscaling

• SeargeUpscaleModels

• SeargeLoras

• SeargePipelineStart

• SeargeImageAdapterV2

• SeargePromptAdapterV2

• SeargeControlnetModels

• SeargeControlnetAdapterV2

• SeargeFreeU

Nodes that have failed to load will show as red on the graph.

Great work, i just cant figure out how to enable Upscaling.

Hires is OK create image with preview. But Upscaling just dont and yes i have set final_upscale_size = 2, but still nothing.

I have this error message when generate an image :
"Error occurred when executing SeargeMagicBox: free_upper_bound + pytorch_used_bytes[device] <= device_total INTERNAL ASSERT FAILED at "..\\c10\\cuda\\CUDAMallocAsyncAllocator.cpp":540, please report a bug to PyTorch."
How can I solve this pleaaaase ? (I have all nodes and models installed)

Is there a way to get clip skip to work with this workflow? Am I just missing it?

Also, like others I was facing some errors when trying to run the workflow. The Windows installer didn't install any of the required models listed in the readme, so I downloaded all of them included the recommended and everything worked fine after that.