NOTE: To use the git clone feature you have to download git, link in description. If you're having issues with using Manager to install missing nodes, make sure you go to Manager and click update all. The red node issue should go away. Topaz Labs: topazlabs.com/ref/2377/
Man, you're the first person I've ever seen communicate how something works. Clear, to the point and simple. Thank you very much. Like and subscribe with bell.
i was using only automatic1111 so the drag and drop a png and you get your settings just blow my mind !!!! I was "what ?! for real, no way..." for almost 5 min. Comfyui look really nice thx for tuto
Great tutorial. ComfyUI is such a powerful and flexible tool, especially for complex automated workflows with dynamic inputs. You can even run your own LLM as part of the workflow. It's the #1 choice for professionals.
This is such a deep dive into Animatediff and the best Animatediff tutorial i have been come across. Thank you so much and off course i liked,..and i was already subscribed. This video deserves many, many views 🙂
NODES!!! yeah even with 3D animation and VFX workflows, nodes were always intimidating coming from a traditional Adobe based background... I used to always joke Nodes = Nose bleeds but once people grasp the power of a nodal workflow you can definitely attain some really in-depth results with visually every aspect of the work created via a mostly non-destructive workflow... I have only dabbled a little with comfyUI but def see the potential despite the fact it will scare away casual SD users... Either way great videos as always keep up the good work my friend...
@@enigmatic_e reminds me of when I first opened Houdini after years of using Cinema 4D... I think in 2024 I will make my 4th attempt at shifting to Blender... I don't use much AI these days but it is good to see more alternatives on the table outside of Automatic1111 for local runs...
Being a newbie with Comfyui...There was a bit of a struggle to get the template to work. Probably due to me already installing comfyui and not following how you did it to be fair. Great tutorial! Being forced to problem solve missing stuff forced me to sort of understand stuff. Still a bit confused and it was probably due to me having both AnimateDiff and AnimateDiff Evolved too. I was thrown a curve ball using 3d Cartoon for SDXL vs the V3 version. Last curve ball I missed was changing the preview window to image / gif vs default h264...I got a mmpeg error with h264. Thanks again for a cool run through tutorial work flow!
Thank you man! I would love if you choose one of those workflows and really dive deep into what every node does and how to tweak it to get different results. But this overview also helped me a lot in terms of basic understanding!
Thanks for this tutorial. This such a stunning place full of so much potential. Thank you for assisting me in my own journey to become a visionary artist.
Great video! You make everything so much more understandable. Comfyui really intimidated me, but now from watching your videos I get why they call it "comfy". Load peoples workflows and just chill haha. Thank you for your amazing work man!
Just had to sub... Content is great and easy to understand. Plus linking all the flows and external resources made everything so simple. Really appreciate you!
Thanks for the tutorial mate, was struggling with comfyUI and really appreciate the informations. thanks for sharing your workflows as well, really, really appreciate it man
Great video! One thing I recall someone testing in the discord is with prompt travel it can often slowly blend. For example if you have prompt A at frame 0 and prompt B at frame 10, it will gradually shift to prompt B over the 10 frames. If you repeat Prompt A and frame 9, it will be a more jarring shift (good for eye blinks and facial expressions etc)
@AIWarper well i tried this as well , problem is it regenerates prompt a on frame 9 , and some parts are different. in my own example there was a girl and father in the scene , girl stayed same but father got aged instantly at frame 9 :) i wish we can fix that problem somehow
@@AIWarper İ think you missunderstood me , dad getting older wasnt something i asked from ai to begin with , on frame 0 it created young dad , but on frame 9 it created older dad (this only happened because i repeated the frame 0 prompt on frame 9 ) So it wouldnt matter which frame i repeated the prompt , it will create another picture different from frame0. Or maybe there is a way to stop this , like using fixed instead of randomize , ill try that tomorrow
Hey man, thanks for the video. Been watching you since early SB solving frame consistency issue, glad that we reach this level of animation now, and u are still the one teaching XD
bro why reacter nod not installing🥲 When loading the graph, the following node types were not found: ReActorFaceSwap Nodes that have failed to load will show as red on the graph.
Hey, I've been playing around with this for a bit. For some reason comfy seems to only be catching the first prompt in the prompt scheduling list as well as the overarching prompt. Any advice for getting it to recognize the other prompts?
when i try and run the run_nvidia_gpu batch file, i get this .\python_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build'.\python_embeded\python.exe' is not recognized as an internal or external command,operable program or batch file. couldnt find anything aboutit on the internet, anybody know a fix?
I have the same issue! i tried so many ways but it's not load my models folder from my SD location. Please i need some help. ComfyUI keep loading models from default folder.
Does the batch size control the over all amount of frames then? What about frames on preview and video output when set to 8? If I want 1280x720 would I input that on the latent image too?
When I try to route my models from a1111 it does not recognize when I load the new 1.6 version. I still have the old version and route to that the same way which has some of the same models and this works. I'm not sure why it wont route to my a1111 v 1.6. Any thoughts?
Good video, I love the animation! is it possible to "mask" say the body from changing so the face only animate etc. using masking or something? #newbie
Thanks for the video! I'm a beginner and this was helpful. One thing I'm uncomfortable with comfy nodes and manager and bringing in other people's workflows. It could contain viruses, keyloggers. Just FYI yall should be very careful with what you're doing. Anyone can create their own node, you're basically just closing your eyes and installing random software. One way to help, but no sure way, is to check the actual source and github repo of nodes, hopefully they have alot of stars meaning likes.
Yes there is a potential danger. I made a video about it recently where I discuss what to look out for. There’s also new features on manager that show you which devs have been activate for more than 6 months.
Issue: After loading the Txt2Img of AnimateDiff into ComfyUI I get the following error: When loading the graph, the following node types were not found: ADE_AnimateDiffLoaderWithContext Nodes that have failed to load will show as red on the graph. Manager shows 'AnimateDiff Evolved' when I click on 'Install missing...', but says it's installed. Solved: Manager > Fetch Updates > Update ComfyUI > Restart. Incase someone has the same problem.
NOTE: To use the git clone feature you have to download git, link in description. If you're having issues with using Manager to install missing nodes, make sure you go to Manager and click update all. The red node issue should go away.
Topaz Labs: topazlabs.com/ref/2377/
Thank you!!!
THANK YOU!! Fixed me right up. I couldn't figure it out
Yeeehhh! This works❤thanks
what about image to animate instead of text? I really need a image to vid one please.
@@RajChandamusic check out some other of my Comfyui videos. There should be a playlist of videos related to Comfyui on my channel.
Man, you're the first person I've ever seen communicate how something works. Clear, to the point and simple. Thank you very much. Like and subscribe with bell.
i was using only automatic1111 so the drag and drop a png and you get your settings just blow my mind !!!! I was "what ?! for real, no way..." for almost 5 min. Comfyui look really nice thx for tuto
This tutorial finally gave me a decent understanding of how ComfyUI works. Thanks so much!!
Great tutorial. ComfyUI is such a powerful and flexible tool, especially for complex automated workflows with dynamic inputs. You can even run your own LLM as part of the workflow. It's the #1 choice for professionals.
What's the module to connect an LLM?
except it's impossible to install
bro i been on comfy for about 2 months and i still learned alot from this video bro thankyou
mm-Stabilized_mid.pth WHERE DO I PLACE THIS PTH? WHAT FOLDER BROTHER?
This is such a deep dive into Animatediff and the best Animatediff tutorial i have been come across. Thank you so much and off course i liked,..and i was already subscribed. This video deserves many, many views 🙂
Glad it was helpful!
Great video, thanks a lot! I knew nothing about comfyUI but this gives me a great sense of how it all works.
excellent tutorial E, really appreciate the time you take to put these together for us.
The quality of your work is incredible! Super good vibe
Appreciate it!
NODES!!! yeah even with 3D animation and VFX workflows, nodes were always intimidating coming from a traditional Adobe based background... I used to always joke Nodes = Nose bleeds but once people grasp the power of a nodal workflow you can definitely attain some really in-depth results with visually every aspect of the work created via a mostly non-destructive workflow... I have only dabbled a little with comfyUI but def see the potential despite the fact it will scare away casual SD users... Either way great videos as always keep up the good work my friend...
Yea it really allows for a lot of more complex setups. I really like it.
@@enigmatic_e reminds me of when I first opened Houdini after years of using Cinema 4D... I think in 2024 I will make my 4th attempt at shifting to Blender... I don't use much AI these days but it is good to see more alternatives on the table outside of Automatic1111 for local runs...
Being a newbie with Comfyui...There was a bit of a struggle to get the template to work. Probably due to me already installing comfyui and not following how you did it to be fair. Great tutorial! Being forced to problem solve missing stuff forced me to sort of understand stuff. Still a bit confused and it was probably due to me having both AnimateDiff and AnimateDiff Evolved too. I was thrown a curve ball using 3d Cartoon for SDXL vs the V3 version. Last curve ball I missed was changing the preview window to image / gif vs default h264...I got a mmpeg error with h264. Thanks again for a cool run through tutorial work flow!
you are a saint for spending your time and energy to share your knowledge
Thank you man!
I would love if you choose one of those workflows and really dive deep into what every node does and how to tweak it to get different results. But this overview also helped me a lot in terms of basic understanding!
What a clean, straightforward tutorial! Great love from South Korea. Looking forward to next videos!!
Thank you! I would love to visit South Korea someday.
Dude. Can I just say, thank you for being you. Such a generous thing you're doing. I totally owe you. Wonderful presentation.
Thanks for this tutorial. This such a stunning place full of so much potential. Thank you for assisting me in my own journey to become a visionary artist.
Thank you for showing the different results from the various animatediff models
Great video! You make everything so much more understandable. Comfyui really intimidated me, but now from watching your videos I get why they call it "comfy". Load peoples workflows and just chill haha. Thank you for your amazing work man!
Glad I could help!
Im watching you for while and I learned a lot and Im grateful that guys like you exist ;) ... all the best have and have good life ;)
Just had to sub... Content is great and easy to understand. Plus linking all the flows and external resources made everything so simple. Really appreciate you!
Dude. I've been obsessed with AI for a little over a year now and your content is the best. Can't wait to dive into animate diff.
Great video thanks a lot broski !! After months of contemplating over A111, finally switched to ComfyUI today after watching your video ! 🔥
Glad I could help!
Awesome video!!!! Thanks.. Spent 6 hours making blurry videos then spent 20 minutes watching this.. fire🔥🔥🔥
Thanks!
No problem 🙏🏽
Shhhh, Don't tell anyone but you're tutorials are my favourites.
great job. Happy to support the channel
thanks for helping me get started. the beast is coming
really well put tutorial ! Will definitely start soon experimenting with Animated Diff
Great job! Very clear and concise instructions. Easy to follow along with. Great work. I look forward to many more tutorials from you.
Glad that you at least asked people to sub and like.!! Rooting for you, its been a really great journey learning for you Enigmatic Sensei..🙌
Thanks for the tutorial mate, was struggling with comfyUI and really appreciate the informations. thanks for sharing your workflows as well, really, really appreciate it man
Glad it helped!
Great video. Got me setup to start creating very easily and provided some great background. Keep up the great content!
Really appreciate the tutorials. thanks for putting these out there
Thanks for making these bro! Great tutorial for Comfyui!
Thank you for this. I have been trying to get over my Comfy UI scare, and this video is very helpful.
Give it a try, it’s not so bad.
you always rocks bro! i was out of my main computer for 2 months going back now and finally trying animatediff thanks to you! 🙏🙏
how far you'e come in 8 months is amazing
Thank you for this great tutorial! A perfectly structured step for step guide!
Thank you for makig this video, I think this will suit the way I work better as I am used to node wranggling in nuke and blender
really love the clarity of your videos, thanks!
Great video!
One thing I recall someone testing in the discord is with prompt travel it can often slowly blend. For example if you have prompt A at frame 0 and prompt B at frame 10, it will gradually shift to prompt B over the 10 frames. If you repeat Prompt A and frame 9, it will be a more jarring shift (good for eye blinks and facial expressions etc)
Oh cool! I didn’t know that’s how it worked. I’ll definitely give that a try! Thanks for sharing that info. 🙏🏽
@AIWarper
well i tried this as well , problem is it regenerates prompt a on frame 9 , and some parts are different. in my own example there was a girl and father in the scene , girl stayed same but father got aged instantly at frame 9 :) i wish we can fix that problem somehow
@@zweiche definitely takes some massaging for sure haha you could try inserting it at say… frame 6, then from 6-10 it would slowly do it
@@AIWarper İ think you missunderstood me , dad getting older wasnt something i asked from ai to begin with , on frame 0 it created young dad , but on frame 9 it created older dad (this only happened because i repeated the frame 0 prompt on frame 9 )
So it wouldnt matter which frame i repeated the prompt , it will create another picture different from frame0.
Or maybe there is a way to stop this , like using fixed instead of randomize , ill try that tomorrow
Thanks for going over the automatic1111 reference. I can use my automatic 1111 models now.
Excelent video and tutorial! Much appreciated for making it!
keep produce this kind of tutorial dude, I learned a lot from it
Thanks!
Such a cool work!! Man its insane!
Thanks for the Excellent intro to ComfyUI !!
Best explanation ever, great, great job!
Waited for this tutorial. worth the wait! Thank you.
🎉
Very informative and useful tutorial, thank you!
Hey man, thanks for the video. Been watching you since early SB solving frame consistency issue, glad that we reach this level of animation now, and u are still the one teaching XD
Yea we’ve come a long way, still a lot to come. 🙏🏽
good pace, no BS tutorial, thanks.
amazing tutorial, very very helpful! Thank you a lot for your work!
Great walkthrough. Gonna give it a try!
Very comprehensive, great stuff!
🙏🏽🙏🏽🙏🏽
So glad I found your channel! Thank you for this video. I'll try it tomorrow.
23:00 Is there a way to maintain the original image's style, while interpolating it from A to B?
bro why reacter nod not installing🥲
When loading the graph, the following node types were not found:
ReActorFaceSwap
Nodes that have failed to load will show as red on the graph.
Did you update everything with manager and did you use install missing nodes?
Keep doing these videos!! You are inspiring lots of people believe me ;)
Looks good. Clear tutorial on how to use some nodes particularly for animations. I guess it's time to start making some.
Nice tutorial and very well explained. keep the good work!
Thanks, will do!
Mate. legendary. Thank you!
clean and wide interpretation. Thnx!
Props! love it. thanks for your hard work.
thank you! I appreciate the guide for comfy and animatediff
You're welcome!
Thanks Fam! Great video
You're a legend. Thank you for this!
No problem!
AnimateDiff is the future of AI video for sure. Love it!
100% agreed!
Well explained, you're a good teacher.
Thank you for teaching me how to fix Node Red .How easy is your explanation?
Mad respect for the effort you put into your videos! #stablediffusion #tutorials
I appreciate that!
Well explained video, very clear. Thank you
Wonderful tutorial. Thanks for sharing ❤
Hey, I've been playing around with this for a bit. For some reason comfy seems to only be catching the first prompt in the prompt scheduling list as well as the overarching prompt. Any advice for getting it to recognize the other prompts?
Great as usual!
Nice work E!
Thanks!🙏🏽🙏🏽
Great tutorial mate cheers !
🙏🏽🙏🏽
Love it. I'll share it with my friends.
i appreciate your teaching style. Thank you!
Thanks for watching!
I'm a simple man. I see enigmatic_e video, I like.
Just a thanks for all the things you've taught us!
My pleasure!
Needed this badly 😅 diff has been a monster to learn
Just a killer tut!!! Thank you very much
when i try and run the run_nvidia_gpu batch file, i get this .\python_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build'.\python_embeded\python.exe' is not recognized as an internal or external command,operable program or batch file. couldnt find anything aboutit on the internet, anybody know a fix?
Thanks for the files, my brother. - God bless you
seriously , need to repeatly watch the video, and pick up knowledges from it , trust me it helps
Grateful bro!
Great video! Thank you for sharing :)
thanks for helping me get this set up!
No problem
Almost no one mentions that ControlNet directory fix. Thanks!
can you do an example using control nets and in painting or masking?
Amazing video G!!
Your page is my study guide
Would love some more warp
5:35 when i click on the cpk_name nothing happens and when i click the arrow it changes to null
any help?
You mean the checkpoint? Do you have any models?
@@enigmatic_e no i don't, the readme says something about stable diffusion 1.5
v1-5-pruned-emaonly.ckpt, should i get that one?
I have the same issue! i tried so many ways but it's not load my models folder from my SD location. Please i need some help. ComfyUI keep loading models from default folder.
Does the batch size control the over all amount of frames then? What about frames on preview and video output when set to 8? If I want 1280x720 would I input that on the latent image too?
Great Content, learned alot from you
When I try to route my models from a1111 it does not recognize when I load the new 1.6 version. I still have the old version and route to that the same way which has some of the same models and this works. I'm not sure why it wont route to my a1111 v 1.6. Any thoughts?
Good video, I love the animation! is it possible to "mask" say the body from changing so the face only animate etc. using masking or something? #newbie
Thanks for the video! I'm a beginner and this was helpful.
One thing I'm uncomfortable with comfy nodes and manager and bringing in other people's workflows. It could contain viruses, keyloggers. Just FYI yall should be very careful with what you're doing. Anyone can create their own node, you're basically just closing your eyes and installing random software.
One way to help, but no sure way, is to check the actual source and github repo of nodes, hopefully they have alot of stars meaning likes.
Yes there is a potential danger. I made a video about it recently where I discuss what to look out for. There’s also new features on manager that show you which devs have been activate for more than 6 months.
@@enigmatic_e thanks!
Issue: After loading the Txt2Img of AnimateDiff into ComfyUI I get the following error:
When loading the graph, the following node types were not found:
ADE_AnimateDiffLoaderWithContext
Nodes that have failed to load will show as red on the graph.
Manager shows 'AnimateDiff Evolved' when I click on 'Install missing...', but says it's installed.
Solved: Manager > Fetch Updates > Update ComfyUI > Restart.
Incase someone has the same problem.
Thanks for making this!