man you are a legend, i have followed many tutorials and wasted many hours, your video is the best, easy to follow guide and the correct links provided was super helpful, thank you. and yes everything worked great, finally!!!
This is a really great workflow. In my tests it performed better than a LoRA in many cases, and so much easier if your character is AI to begin with. Thanks for sharing! Would be great to see combined with a pose solving solution, to better control output. It can be tricky in prompting to turn the characters head, although when you do, the consistency is quite amazing. Subscribed! :)
My dumbass thought that was a pic of Riley Reid lolllll. Thank you for the video. I actually decided to use ComfyUI for the first time and it took an hour to install everything but this workflow works. Keep up the good work!
I got this issue: "SamplerCustomAdvanced Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! (when checking argument for argument weight in method wrapper_CUDA__native_layer_norm)" Please help!
I keep getting the following error ........ [Errno 13] Permission denied: 'E:\\ComfyUI_windows_portable\\ComfyUI\\models\\mediapipe\\selfie_multiclass_256x256.tflite' The file is there but evidently I need to give some sort of permission for comfyui to access it????
I had that problem until I git-cloned all the models from the ComfyUI_LayerStyle models link to a temp folder (22 folders 80GB) and then moved them to ComfyUI models folder. It's a lot but you can as a bonus run all their example workflows as you will then have all the models for ComfyUI_LayerStyle.
Hi, I always see missing node error. I tried cloning both Pulid-Flux & Pulid-Flux-Enhanced in custom-nodes folder manually also to see if the error gets resolved but not able to load "Pulid Flux Loader Model, Pulid Eva Clip Loader, insightface loader, apply pulid flux"
I was using portable version and I think that was getting me this error. Now everything is working fine but I get an error on "SampleCustomAdvanced" that says "forward_orig() takes from 7 to 9 positional arguments but 10 were given" ... I have RTX 4060. I tried using t5 Q4 as well but not sure what this error is.
ERROR: Could not install packages due to an OSError: [Errno 13] Permission denied: 'C:\\Python311\\Scripts\ umba' Consider using the `--user` option or check the permissions. What does this error mean? i get this when trying to install the requirements.txt
Does this only work with the portable version? I have downloaded more than 20 workflows, I have watched more than 30 videos and none of them have worked for me, this is one more, thousands of different errors without a solution, I am already losing hope of making comfyui work
Hi, thanks for the video, I actuallty tried to refine flux images with SDXL too, but I just felt that it was just better to generate with only SDXL and upscale it. May be I didn't notice something that you did? Do you feel there is any advantage generating the image with flux first ? other than getting better hands with flux offcourse
It's more about the coherency, if you create normal portraits then SDXL will be cool, but if you want something more complex (maybe tattoo design, jewellery shape or color) etc then SDXL loses the coherency 😶...It is mostly depending on the use case
The latest ComfyUI portable comes with python 3.11.9 and so it appears that I would need to downgrade to 3.8. "ERROR: Package 'facelib' requires a different Python: 3.11.9 not in '=3.5'"
@@xclbrxtra I tried that first but after error "python setup py egg_info did not run successfully etc., ... ModuleNotFoundError: No module named 'filterpy'" I then followed a forum recommended for portable comfyui facexlib installation issues to download 'facelib-1.2.1-py3-none-any whl' and directly install that. That's when I got the Package 'facelib' requires a different Python error. PS: In hindsight it looks like maybe facelib instead of facexlib is possibly incorrect at least with the direct method. I will check....
Dear, Thank for great video. I follow every step but my ComfyUI (portable) still cant recognize PULIDFLUX node, Im sure the insightface has properly installed
@@xclbrxtra Yes, I have. I run all the command for insightface , Facexlib and the requirement.txt again and again. And it said all files are already installed. model, clip and vae are in place. Can my GPU be problem? I use rtx 3080 10gb. Many thanks
@@xclbrxtra Holy shit, somehow I fixed it ( i still dont know that work, lol ) I installed PuLID (NeurIPS 2024), miniconda, VS2015, VS2023 then fix the PuLID_ComfyUI in the Manager and it just work. anyway Thank you again for great video
Seems you are new to AI Image generation. The only other way to do it in generative AI is with Lora and it needs resources and time to train just 1 face.
LayerMask: PersonMaskUltra V2 [Errno 13] Permission denied: 'C:\\SD\\Comfyui\\ComfyUI_windows_portable\\ComfyUI\\models\\mediapipe\\selfie_multiclass_256x256.tflite' Not able to get past this part, as I'm given this error, does anyone have a fix for this issue? Please help.
Error occurred when executing DualCLIPLoaderGGUF: module 'comfy.sd' has no attribute 'load_text_encoder_state_dicts' File "C:\COMFY-UI\ComfyUI\execution.py", line 152, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\COMFY-UI\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:\COMFY-UI\ComfyUI\execution.py", line 75, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\COMFY-UI\ComfyUI\custom_nodes\ComfyUI-GGUF odes.py", line 372, in load_clip return (self.load_patcher(clip_paths, clip_type, self.load_data(clip_paths)),) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\COMFY-UI\ComfyUI\custom_nodes\ComfyUI-GGUF odes.py", line 327, in load_patcher clip = comfy.sd.load_text_encoder_state_dicts( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Any idea what does it mean?
PermissionError: [Errno 13] Permission denied: 'C:\\ComfyUI_windows_portable\\ComfyUI\\models\\mediapipe\\selfie_multiclass_256x256.tflite' can't find a way to make that work
This is easily the best channel for people trying to run Comfyui with limited VRAM. The info here is nothing short of amazing. Thank you sir
Thank you so much! Your install guidence solve a big promblem!
man you are a legend, i have followed many tutorials and wasted many hours, your video is the best, easy to follow guide and the correct links provided was super helpful, thank you. and yes everything worked great, finally!!!
you left out the important part of explaining the actual workflow items
This is a really great workflow. In my tests it performed better than a LoRA in many cases, and so much easier if your character is AI to begin with. Thanks for sharing!
Would be great to see combined with a pose solving solution, to better control output. It can be tricky in prompting to turn the characters head, although when you do, the consistency is quite amazing. Subscribed! :)
Yes, I am working on a workflow to integrate controlnet and this together so posing can be controlled, I'll share a video soon 💯
@@xclbrxtra Fabulous! I have also tried to include Lumiere, instead of the SDXL detailer. Keep up the good work!
This works! Thank you for all the extra explanations!
My dumbass thought that was a pic of Riley Reid lolllll.
Thank you for the video. I actually decided to use ComfyUI for the first time and it took an hour to install everything but this workflow works. Keep up the good work!
I got this issue: "SamplerCustomAdvanced
Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! (when checking argument for argument weight in method wrapper_CUDA__native_layer_norm)" Please help!
sam error i got have you cleared
I keep getting the following error ........ [Errno 13] Permission denied: 'E:\\ComfyUI_windows_portable\\ComfyUI\\models\\mediapipe\\selfie_multiclass_256x256.tflite' The file is there but evidently I need to give some sort of permission for comfyui to access it????
same
I had that problem until I git-cloned all the models from the ComfyUI_LayerStyle models link to a temp folder (22 folders 80GB) and then moved them to ComfyUI models folder. It's a lot but you can as a bonus run all their example workflows as you will then have all the models for ComfyUI_LayerStyle.
same here
got same error unfortunately...🙁
U need to give to all permissions to the folder in use, check log what's going on there, i have same but after solved,
is there any way to add controlnet on this workflow? I've tried a lot but I couldn't make that yet.
I did everything as advised but keep on getting the " UnetLoaderGGUF
too many values to unpack (expected 2) " error. running on a 3060
have a problem with pulid flux enhanced , impossible to make it work, impossible to fix
Hi, I always see missing node error. I tried cloning both Pulid-Flux & Pulid-Flux-Enhanced in custom-nodes folder manually also to see if the error gets resolved but not able to load "Pulid Flux Loader Model, Pulid Eva Clip Loader, insightface loader, apply pulid flux"
I was using portable version and I think that was getting me this error. Now everything is working fine but I get an error on "SampleCustomAdvanced" that says "forward_orig() takes from 7 to 9 positional arguments but 10 were given" ... I have RTX 4060. I tried using t5 Q4 as well but not sure what this error is.
Thx man, useful vid!
i got this error (SamplerCustomAdvanced) forward_orig() takes from 7 to 9 positional arguments but 10 were given
any help, please
Hi, I have the same problem. Please tell me if it was solved?
ERROR: Could not install packages due to an OSError: [Errno 13] Permission denied: 'C:\\Python311\\Scripts\
umba'
Consider using the `--user` option or check the permissions. What does this error mean? i get this when trying to install the requirements.txt
Does this only work with the portable version? I have downloaded more than 20 workflows, I have watched more than 30 videos and none of them have worked for me, this is one more, thousands of different errors without a solution, I am already losing hope of making comfyui work
Hi, thanks for the video, I actuallty tried to refine flux images with SDXL too, but I just felt that it was just better to generate with only SDXL and upscale it. May be I didn't notice something that you did? Do you feel there is any advantage generating the image with flux first ? other than getting better hands with flux offcourse
It's more about the coherency, if you create normal portraits then SDXL will be cool, but if you want something more complex (maybe tattoo design, jewellery shape or color) etc then SDXL loses the coherency 😶...It is mostly depending on the use case
The latest ComfyUI portable comes with python 3.11.9 and so it appears that I would need to downgrade to 3.8.
"ERROR: Package 'facelib' requires a different Python: 3.11.9 not in '=3.5'"
Is this when you try pip install facexlib ? Because I too have python 3.11 and it worked, I have not downgraded 😶
@@xclbrxtra I tried that first but after error "python setup py egg_info did not run successfully etc., ... ModuleNotFoundError: No module named 'filterpy'"
I then followed a forum recommended for portable comfyui facexlib installation issues to download 'facelib-1.2.1-py3-none-any whl' and directly install that. That's when I got the Package 'facelib' requires a different Python error. PS: In hindsight it looks like maybe facelib instead of facexlib is possibly incorrect at least with the direct method. I will check....
@@xclbrxtra Finally I got mine working. I had to include '--use-pep517'
Summary (i.e., from python_embeded):
python -m pip install --use-pep517 facexlib
python -m pip install insightface-0.7.3-cp311-cp311-win_amd64.whl
python -m pip uninstall onnx onnxruntime
python -m pip install onnx==1.16.2 onnxruntime==1.19.2
python -m pip install onnxruntime-gpu==1.19.2
python -m pip install ftfy
python -m pip install timm
@@bwheldale I have same issue! It's frustrating!
I think it has to do with the portable python embedded version
Great video!! Do you have a good img2img workflow for Flux Comyui?
Best consistent character tutorial, thank you!
Great video! Thank you very much!!!
Hardest installer Custom node for me...I cant install
Dear, Thank for great video. I follow every step but my ComfyUI (portable) still cant recognize PULIDFLUX node, Im sure the insightface has properly installed
Did you put the antelopev2 files in models/insightface/models/antelopev2?
@@xclbrxtra Yes, I have. I run all the command for insightface , Facexlib and the requirement.txt again and again. And it said all files are already installed. model, clip and vae are in place. Can my GPU be problem? I use rtx 3080 10gb. Many thanks
@@xclbrxtra Holy shit, somehow I fixed it ( i still dont know that work, lol ) I installed PuLID (NeurIPS 2024), miniconda, VS2015, VS2023 then fix the PuLID_ComfyUI in the Manager and it just work. anyway Thank you again for great video
Is there any way to achieve the same with replicate?
Great Job my man! Thank you!!!
sen adamsın. senin tassagini yerim.
so finally he can do it with flux, what we allready can do 1000times better and faster... but he is using flux 😂
Seems you are new to AI Image generation. The only other way to do it in generative AI is with Lora and it needs resources and time to train just 1 face.
excelent, works fine, but so slow..
LayerMask: PersonMaskUltra V2
[Errno 13] Permission denied: 'C:\\SD\\Comfyui\\ComfyUI_windows_portable\\ComfyUI\\models\\mediapipe\\selfie_multiclass_256x256.tflite'
Not able to get past this part, as I'm given this error, does anyone have a fix for this issue? Please help.
Bhai dm dekho yaar
Error occurred when executing DualCLIPLoaderGGUF:
module 'comfy.sd' has no attribute 'load_text_encoder_state_dicts'
File "C:\COMFY-UI\ComfyUI\execution.py", line 152, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\COMFY-UI\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:\COMFY-UI\ComfyUI\execution.py", line 75, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\COMFY-UI\ComfyUI\custom_nodes\ComfyUI-GGUF
odes.py", line 372, in load_clip
return (self.load_patcher(clip_paths, clip_type, self.load_data(clip_paths)),)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\COMFY-UI\ComfyUI\custom_nodes\ComfyUI-GGUF
odes.py", line 327, in load_patcher
clip = comfy.sd.load_text_encoder_state_dicts(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Any idea what does it mean?
PermissionError: [Errno 13] Permission denied: 'C:\\ComfyUI_windows_portable\\ComfyUI\\models\\mediapipe\\selfie_multiclass_256x256.tflite'
can't find a way to make that work