You are amazing! I've been looking for a creator like you who cares to teach people and starts every workflow from scratch and explains to logic. Seriously thank you so much. Cant wait to watch the rest of your videos and learn from you
Yeah I'm trying to go in real depth. It's VERY time consuming, but it's worth it. I have 3 videos in the works already, one beginner, one middle, one advanced - that way everyone will be happy (or not one will be happy hahaha) . Thank you so much!!!!!! 👽
Yeah your account might not rank well. I had to remember your exact name to find you. Well exactly name and video lol. Anyhow I loved this video. You're a great teacher. Looking forward to more. @@stephantual
Thanks for this really cool video. I found your channel from a comment that you made on Nate Black’s channel. I assume that since you’re watching those videos that you want to grow your channel and I wish that I had people’s stream of consciousness thought on why they clicked on my video and why they clicked away so I thought I would do that for you. I clicked on it because I was intrigued by how using the software could improve my videos on my channel. However, since I’m not really your typical target audience, I was immediately lost and realized that I needed to look up something that’s more for beginners, so I stopped the video after only being a couple minutes in so that I can look up a video that explains what ComfyUI is. I thought I’d pop this coming in before navigating away, though. Good luck to you on your channel. ♥️
Cheers! That's good feedback. I realize what I do is very, very , very niche but ... that's just the way I'm built! 😅 Since I was 7 I was fascinated by computers, the demoscenes, etc - I was coding before I could ride a bike haha! I'm considering making 'new user' friendly content, perhaps a series or something like that, to learn the basics, and more importantly, the underlying principles. Thank you for your feedback! 👽
@@MamaRosie Hey are you on discord? Would be fun to chat about the stuff that's not often talked about when making small niche creator content with passion :)
Great tutorial. My Ipadapter is installed but the "IPadapter Apply" is missing. Do I have to reinstall the custom node IPAdapter_plus? Everything is working fine on workflow.
Mostly, people from comfyUI tutorial tries to describe it like some some form of art: canvas, paint, etc... Which is more confusing... It is much more sense to describe comfyUI like programing (you)... While me, I see comfyUI like flowchart in production... I think people are secretly have that production mentality thats why some people just love comfyUI. It just that they're unconscious to it and can't describe properly...
Hello and thanks for the video! Do it exist a easier wotkflow, this one is huge for a newbie as I am :), just to make face skin more realistic and less plastic? Just an img2img to add detail and do a skin more realistic :)
Hi and thanks for the heads up! Which link is down? I'm currently updating all my worfklows and will push them one by one to my profile at flowt.ai/community?search=%3Apub%20Stephan_Tual as they support more nodes and have better support. Let me know and I'll fix it! Cheers! 👽
I have this problem, can you help me? D:\Artificial_Intelligence\ComfyUI\python_embeded\Lib\site-packages\timm\models\vision_transformer.py:91: UserWarning: 1Torch was not compiled with flash attention. (Triggered internally at ..\aten\src\ATen ative\transformers\cuda\sdp_utils.cpp:263.) x = F.scaled_dot_product_attention( ERROR:root:!!! Exception during processing !!! ERROR:root:Traceback (most recent call last):
Bro join the discord it's impossible to debug anything in YT comments 👽Looks like you're missing xformers or that your pytorch is out of date, or maybe you don't have the requirements installed. See you soon! 👾
hmm... the results are to different to the original image. U are not only upscaling, you are changing the mood, the content and details of the images. for me Its importent to keep details of an image and keep the details. Otherwise you can generate a new image directly without tons of nodes.
Looks like CF was down for a day or so. I'm currently rewriting my flows to support the new stuff (new supir, new IPA, better CNs, etc) - will upload everything to my profile at flowt.ai/community?search=%3Apub%20Stephan_Tual as they are more stable. Thanks for the heads up, I really appreciate it! 🙏🏻👽
I am very new to this and don't understand 1/10th of what you are showing in the video, but I will eventually. I haven't been this excited about tech, since VMWare changed the datacenter with server virtualization.
Good question. Most of the underlying algos and programs 'wrapped' by nodes in comfyUI are available in A1111. To be clear there's no 'favoritism' on my part, my view is and has always been clear: the only thing that matters is that you feel 'at home' with the tools and to create great, engaging content! Cheers! 👍👽
not cheer but sad personally i dont like comfy UI stpd tool bcz its less powerful amazing and rich then A1111, A1111 is leading in AI local tool everything is inside it ii bet if you want you can make a tutorial on this tool inside of A1111 in next video I will really feel comfy then and share it too will wait deer :)@@stephantual
hello, thanks for the tutorial. when i load the workflow, and install the missiging nodes, i get the following error: When loading the graph, the following node types were not found: CCSR_Model_Select CCSR_Upscale The nodes that could not be loaded will appear in red in the graph. I have tried to do it with the direct link and I still get the same thing. Any suggestions? Thank you very much.
Mmm... just checked and both nodes install correctly through both GH and the manager. Can you please use the discord (tinyurl.com/URSIUM) please as RU-vid comments aren't really good to post code etc 😄 Cheers! 👽