Awesome tutorial! And Love seeing you're working with clear flows like a professional rather than trying to pack nodes as tightly as possible. Great stuff
You're welcome, and that's a good observation. I'm using the older PulId model here because it’s highly compatible with FLUX. especially when dealing with complex inputs. However, the newer model offers exciting improvements!
hello! I have been trying to solve this error and I have not been able to: KSampler 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)
you need to download the zip (.PY) file, then replace into the - custom_nodes\ComfyUI-PuLID-Flux-Enhanced use the link here github.com/sipie800/ComfyUI-PuLID-Flux-Enhanced/issues/54 then scroll all the way down in the forum. Follow the ⭐TLDR instruction.
i get this error while fixing facexlib, what should i do? "AttributeError: module 'pkgutil' has no attribute 'ImpImporter'. Did you mean: 'zipimporter'? [end of output] note: This error originates from a subprocess, and is likely not a problem with pip. error: subprocess-exited-with-error"
Hi there, I apologise for reading this. Attempt to use the console log submit or find issue feature. There might be a solution offered if someone else has posted a problem like that in the forum. Additionally, if you are using a virtual environment, make sure that dependencies are not conflicting and that the environment is correctly activated.
How do I install this on ComfyUI Desktop, previously known as V1. It's a beta & the file structure is different than yours. There is no Python-embedded folder or Python file embedded in the structure. Can I download an external copy of Python & use that? Then I won't know the path to install insightface or facelib. Why can't I use the models that are in the manager?
Hello there, Thanks for bringing this up! I haven’t worked with the new ComfyUI Desktop (V1 beta) yet, but I’ll definitely look into it and explore how the new file structure works. I’ll check if the beta version changes how these models are integrated.
Bearded faces can be tricky! you are right, however you can emphasize the beard details in your input prompts, like “thick brown beard” to guide the model better. secondly inpainting can clean up or enhance the beard if it doesn’t come out as expected.
Because this workflow is simplified and does not use the diffuser node, ensure that the model file you are attempting to load is compatible with the version or UNET you are using.
when i generate.. CheckpointLoaderSimple ERROR: Could not detect model type of: D:\XXX\ComfyUI_windows_portable\ComfyUI\models\checkpoints\flux1-dev-fp8.safetensors
You must use and download the FP8 checkpoint model for FLUX. You can see the link to the guide here. th-cam.com/video/TWSFej_S_bY/w-d-xo.htmlsi=jisA9EJyon1W-CHv
Awesome tutorial! And Love seeing you're working with clear flows like a professional rather than trying to pack nodes as tightly as possible. Great stuff
Your feedback is greatly appreciated. Thank you for your thoughts and encouragement
woow. dude this is helpful.! top man! cheers from Estonia!
Thank you so much! Cheers to you all the way in Estonia! 🌍
感谢你，终于再次看到了你得更新
It is nice to read from you, and you are very welcome.
Muito obrigado por compartilhar. Isso vai me ajudar muito. Tudo de bom meu amigo :)
Thank you so much for your amazing comment! Tudo de bom para você também, meu amigo!
Thx! 😊
Why are you using the older PulId model?
yeah there is a new one available
You're welcome, and that's a good observation. I'm using the older PulId model here because it’s highly compatible with FLUX. especially when dealing with complex inputs. However, the newer model offers exciting improvements!
You are right, you'd like to try it, and feel free to experiment as well
Your video easy to learn ❤❤
Thank you for sharing your experience.
i dont have the "POWER LORA" like 06:15 // am i miss somthing?!😑
To use the Power Lora node, first install the custom node RGTHREE by using the manager.
hello! I have been trying to solve this error and I have not been able to:
KSampler
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)
you need to download the zip (.PY) file, then replace into the - custom_nodes\ComfyUI-PuLID-Flux-Enhanced
use the link here github.com/sipie800/ComfyUI-PuLID-Flux-Enhanced/issues/54
then scroll all the way down in the forum.
Follow the ⭐TLDR instruction.
i get this error while fixing facexlib, what should i do?
"AttributeError: module 'pkgutil' has no attribute 'ImpImporter'. Did you mean: 'zipimporter'?
[end of output]
note: This error originates from a subprocess, and is likely not a problem with pip.
error: subprocess-exited-with-error"
Hi there, I apologise for reading this. Attempt to use the console log submit or find issue feature. There might be a solution offered if someone else has posted a problem like that in the forum.
Additionally, if you are using a virtual environment, make sure that dependencies are not conflicting and that the environment is correctly activated.
I thank you for turtorial, I setup is working. I test another channal video is not working. You video is best 🙂
Thank you so much for letting me know! I'm glad you found what you were looking for
How do I install this on ComfyUI Desktop, previously known as V1. It's a beta & the file structure is different than yours. There is no Python-embedded folder or Python file embedded in the structure. Can I download an external copy of Python & use that? Then I won't know the path to install insightface or facelib. Why can't I use the models that are in the manager?
Hello there, Thanks for bringing this up! I haven’t worked with the new ComfyUI Desktop (V1 beta) yet, but I’ll definitely look into it and explore how the new file structure works. I’ll check if the beta version changes how these models are integrated.
@goshniiAI Thanks. Hopefully soon, because I'd really like to use this, I also watched @Mickmumpitz video & ran into the same situation.
@@AlfaOmegaGrafxthe python executable that you need to use is .venv/Scripts/python.exe or pip.exe (or without .exe on unixoid OSs)
I use lighting ai can i do this in lighting ai??
You certainly can. Be sure to enter the necessary parameters for a lighting model.
All these workflows struggle with bearded man image.
Bearded faces can be tricky! you are right, however you can emphasize the beard details in your input prompts, like “thick brown beard” to guide the model better. secondly inpainting can clean up or enhance the beard if it doesn’t come out as expected.
🦾🦾
💚💚
Hello I need help I’m stuck and I have a headache with this issue
Cannot load
Because this workflow is simplified and does not use the diffuser node, ensure that the model file you are attempting to load is compatible with the version or UNET you are using.
@@goshniiAI can you help me using anydisck
when i generate..
CheckpointLoaderSimple
ERROR: Could not detect model type of: D:\XXX\ComfyUI_windows_portable\ComfyUI\models\checkpoints\flux1-dev-fp8.safetensors
You must use and download the FP8 checkpoint model for FLUX. You can see the link to the guide here. th-cam.com/video/TWSFej_S_bY/w-d-xo.htmlsi=jisA9EJyon1W-CHv