[Edit:
Version v5.0 works with latest comfyui (v0.15.0).
If you have any problems, please refer to the FAQ at the bottom of the page or have a look in the comments.
Many thanks to everyone who tested this workflow. Thank you very much for the many inquiries and, of course, for all the knowledge and experience you have contributed. here👍🙂
Special thanks to:
@SeoulSeeker for the "Dead Simple MMAudio" workflow wich are the basis of the audio part here,
@taek75799 for the really well working enhanced models
@Bakazaya pointing to the color issue in version v3.0 and running lots of tests,
@bluntfeather sharing latest experiances with installing Comfyui-Easy-Install,
@nitrovtx for remain persistent in matters of quality and running a lot of tests,
@Icey64 for providing the link to "Comfyui-Easy Install",
@boinobin730 for asking for a First to Last Frame option, running pre tests and responding fast as hell 🙂 and
@SnowShoes311 thank you so much again for all your buzzing 😋]
Features:
Optimized Wan 2.2 workflow, runs perfect on RTX 3060 12 GB VRAM GPU and 32 GB RAM,
"Text to Video", "Image to Video" and "First/Last Frame 2 Video" generation in one workflow, all with easy audio generation,
easy installation/model downloading, all necessary sources are specified,
easy to use workflow, clearly structured, all necessary steps are explained,
easy switches for mode selection,
easy prompt selection for fast prompt creation/testing,
easy switching between "standard" and "enhanced" models,
very fast and smoth high quality outputs up to aprox. 1440 x 960 with 60fps,
2x fast upscaler,
4x fast framerate multiplier,
MMAudio Sampler (generates sound accordingly to the video action),
Triton and Sage Attention option,
A 5 Second long high quality video generation takes about 10 - 15 minutes (see below).
Tested generation times:
As a rough guide value for RTX 3060 GPU: generating a 5 second long high quality 1440 x 960 60 fps video with 6 steps it will take:
t2v: around 10 - 12 minutes,
i2v: around 15 minutes.
Comfyui-Easy-Install with Triton + SageAttention:
This workflow should work with any latest comfyui version >v0.6.0 (Desktop, Embedded, Windows/Linux).
However, comfyui is developing rapidly, and it often happens that some of the custom nodes used are not updated quickly enough or not updated at all. Manual workarounds are sometimes necessary. Furthermore, care must be taken to ensure that there are no conflicts with other nodes.
If you're having difficulties with your existing comfyui system or if you want to run video generation on a separate (parallel) comfyui system, like I do, I would recommend you the following installer: https://github.com/Tavris1/ComfyUI-Easy-Install.
Complete installation of comfyui including manager and some pre configured custom nodes is just one click - really 🙂
Installation of Triton + SageAttention is just a second click - really 🙂 And since it's so easy now, I would definitely recommend it to you for video generation.
Cause it is an embedded version, you can install it parallel to your existing comfyui version without the risk to ruin your working system.
After installation just configure the "extra_model_paths.yaml" file to use your existing models.
After a fresh installation of Comfyui-Easy-Install you might have some issues too, but there are known workarounds - please see the FAQ below.
For testing/understanding/experimenting/changing the workflow:
Click "Toggle Link Visibility" to see the links.
click the Subgraph symbols to open the Subgraphs.
for quick testing you may lower the settings for: steps, clip lenght and video resolution,
be really carefull with modifying Groups or Subgroups (even Titel or Color) cause they are essential for switching,
feel free to try and test other models. Just give me a hint if you find models which deliver better results and fitting the 12 GB VRAM limit.
And as usual: Have Fun 🙂🙂
Short Conclusion:
This workflow is based on elements of a variety of allready published workflows. My "job" was only to put things together, optimize it for a small machine and create a most simple and hopfully user or even "beginner" friendly workflow.
I`m not an "expert" - just a user who wants to get it running on "available" hardware.
There are many things I don't really understand. If you find mistakes or better solutions please give me a hint.
And I really hope that even "beginners" have a chance to go the first steps...
Frequently Asked Questions (FAQ):
For quick and better overview I will try to merge all known issues here - step by step (please be patiant). If your issue is not listed here, please have a look in the comments first. Most issues have been allready discussed.
Comfyui Nodes 2.0:
Turn off Nodes 2.0 in comfyui (use comfyui menue). Actually not all custom nodes are supported.
Comfyui crashes after generation while vae decode, upscaling or frame rate multiplying (Rife VFI) without any error report:
This is a RAM problem (not VRAM). Increase your swap file (min. 64 to 128 GB) or set it to automatic management on a fast drive with at least 100 GB free space.
JW Nodes (JWFloatToInteger, JWIntergerDiv, JWImageResizeByLongerSide), soundfile missing:
For the workaround look here and here:
python -m pip install soundfileFresh Comfyui-Easy_Install Installation (missing soundfile and Pytorch v2.9.0 issue with SageAttention on Windows:
For full conversation look here.
Open cmd in python_embedded folder:
python -m pip install soundfile python -m pip uninstall -y torch torchvision torchaudiopython -m pip install torch==2.8.0 torchvision==0.23.0 torchaudio==2.8.0 --index-url https://download.pytorch.org/whl/cu126Slider Nodes - how can I modify the "default" values:
Right click the slider node, choose Properties and set the values you like 🙂🙃
Description
PianterI2VAdvanced: "providing dynamic enhancement and intelligent color drift correction". The advanced node improves color "stability" for I2V generation significantly.
some minor cosmetics
FAQ
Comments (29)
Thank you so much for continuing to develop this WF and sharing it. It works great!
Hi - thank you so much for your feedback and buzzing 🙂
Hi Arkinson. Happy New Year, friend. I hope you are doing well in the new year. I have been busy migrating over to a standalone comfyui portable. Stability Matrix became over complicated when i went to 8.2 comfyui. So after migrating over i have tried to get your new workflow working but i am having trouble as i keep getting a blank image as an output. Comfyui 8.2 python is 3.12.10, pytorch is 2.91+Cu126. Do you have any hints what i might be able to do to get it working? thanks Arkinson.......
Nevermind. I got it to work. Since this is a new comfyui setup, I have never run the workflow on the new Comfy setup. First thing I did was try I2V as that is what I use most of the time and it didn't work. I wrote this message to you and tried T2V and it worked. then I did an I2V again and it worked. Just thinking about you channelled positive ComfyUI karma to the workflow. :-)
@boinobin730 Hi - wellcome back and a Happy New Year too 🙂I started the first days of the new year completely out of brain with a nasty cold 😰 Nice to "see" you again 🙂
The latest workflow v3.1 should work now in all cases. The most tricky part were the swich options... The community input here is amazing - so it is a lot of fun to keep it running.
I deleted "StabilityMatrix" after testing over a longer time range, The idea is not bad but in reallity it was some of the most "instable" and buggy software I ever tested. For comfyui I really would advice you "Comfyui-Easy-Install" (have a look in my new model desription and FAQ here). It is the most easy way to install Triton + SageAttention too (just one mouse click). It has become my "favorite" installation for video generation and as a test invironment.
@arkinson Sorry that you got sick around New Years. Getting sick on holiday times, I hate when that happens. I definitely will, regarding triton + sage attention. (triton and Sage not installed yet) I did install the basics from scratch but there was a lot of mucking around and consulting Chatgpt. I got sick of the Comfyui Stability matrix, it blew up too much after every comfy update. I end up spending more time fixing comfy than using comfy. I still use onetrainer in Stability Matrix, finally cracked the PONY lora making now. My character is finally stable, I will now make more characters. LOL.
@boinobin730 You should install Onetrainer from Github too. On Windows it is just to install Python, then Git clone and one click at the bat file. This is very easy - no ChatGPT needed 😉
And once again 😅 I would recommend you a parallel comfyui installation for video generation with the mentioned Easy installer - nothing easier then that.
Btw. I`m on the way to prepare a new simple all-in-one workflow for Wan + sound (InfiniteTalk). Very first quick-and-dirty tests looking curious - if you like have a look at my video gallery.
@arkinson Great idea, I will, since I am slowly getting better with Comfy and general installations.
Wow. this sounds very interesting. I did use a workflow for infinite talk but that was a while ago. I'm always looking forward to what you produce.
@arkinson Fantastic. trying it out right now. Thanks Arkinson
@boinobin730 And here I have pulished version 4.0 (with sound) now 🙂 It simply generates sound to the generated video. It is very easy to use and generates pretty cool sound effects - especially for nstw....
@arkinson This is a good workflow. I ended up changing the NSFW sound model for the standard v2 fp16 model from here. I got a lot of random moaning. https://huggingface.co/Kijai/MMAudio_safetensors/tree/main . I haven't been running these workflows for a while and I think it was mentioned a while ago that the model has a tendency to try to go back to the first frame. Evidenced by the car driving off example. Also I tried to get the woman to go under the waterfall and she is constantly turning back to face the viewer. Was there ever a fix for this peculiarity? I haven't tried any NSFW x rated stuff yet. I will give it a go in time. I just wanted to see what it was capable of. It's not bad at all. Not perfect but for a bit of free sound effects that match with the video, I can't complain.
@boinobin730 Oh my - basketball with moaning 😆🤣 I love this sh**t 😂
Thank you for linking the Kijai models. I will try them too. Never thought that we get too much moaning ín nsfw 🙄🤣 To be honest, I just grabbed the models from seoulseekers workflow.
Sorry, I did not get you: "that the model has a tendency to try to go back to the first frame". Yes, this "looping" movement is the known problem with the "video" models. I tried to reduce it by using the "enhanced" I2V models and the "painter" node at version v3.0 and later. In v4.0 I only added the mmaudio part, wich has absolutely no influance to the gererated video. But maybe I got you completely wrong - If not, please look here again 😉 Yes, I know - everything is getting a little bit confusing more and more. Sometimes I have to pay attention too, on which workflow I'm actually working on. I hope, the new workflow "titles" are better understandable.
@arkinson Yeah. That's the thing I am noticing about video clips. When given enough time the motion tends to be circular. So the car example instead of just driving up the street it morphs into turning around and driving back, like something out of game. I think I will adjust the length of video clip, so that it doesn't have time to circle back. Nothing we can do about it, since it is a model problem. It's all good. You are doing great work here. You are probably one of the most engaged creators on Civitai. A lot of creators aren't vocal at all.
Hi, I have a problem tying to use the workflow, the Unet Loader GGUF And PainterI2VAdvenced are rounded in red and there is a mesage that it is not instaled but i click the button to install and it seems like it does nothing.
If i trie to execute i have this error mesage:
Prompt execution failed
Cannot execute because a node is missing the class_type property.: Node ID '#149'
What can i do?
Make sure you have selected YOUR local model. Right click in the Unet model loader and select the right model from your local path. If you can`t find your model there you might saved it in a wrong path or your extra_model_paths.yaml file is configured wrong.
@arkinson I finally got it working, comfyUI told me it was instaled bit it wasent so i did it manually and it worked.
thank you.
@robertobs299677 I've got the same error with PainterI2VAdvanced and can't get it to work, how did you install it manually? Everything else in the workflow appears good. I'm new to ComfyUI so my apologies if this is a dumb question
@johninvrvam853 What do you mean with "same error with PainterI2VAdvanced" and "can not get it to work"???? Give an exact description of your problem with all necessary informations and steps you allready have done and I may try to help you.
@arkinson The PainterI2VAdvanced node is the only one on the workflow that's bordered in red, and when trying run the workflow I get the error: Cannot execute because a node is missing the class_type property.: Node ID '#260:259'. Also when looking at the PainterI2VAdvanced node instead of having the different adjustments listed like (Width, length, height, batch size, etc.) it just says UNKNOWN for all of those. I've ran the ComfyUI manager which updated and fixed all the errors and missing nodes I had initially, but for some reason PainterI2VAdvanced just doesn't seem to work. If there is a way to manually install this node I'd love to try it, but admittedly I'm new to ComfyUI and this is quite challenging. Thanks for any and all help you can provide!
@johninvrvam853 Hi - thank you 👍 For a comfyui beginner, it is quite challenging ("sporty") to start with video generation 😉🙂 Ok - if I got you right, the custom node PainterI2VAdvanced is not installed or missing on your system. It makes no sense to "run" a workflow with missing nodes. Thats the reason I was confused about your error messages.
Manually installing the node should be the very last try. So first let`s check what you get from the manager. Open the manager -> Custom Nodes Manager -> search for "PainterI2VAdvanced". Do you see the node? Is it installed? Any conflicts?
Btw. wich comfyui version do you use? As a beginner I would really advice you to use Comfyui-Easy-Install, especially for video generation, see my model desription.
@arkinson Thanks for the advice, I may just start over with a clean install. I've been swamped with work so haven't had a chance to mess with it lately, I'll update this weekend any results!
@arkinson So I got PainterI2VAdvanced working and no red bordered nodes when your workload is displayed, so all good there. Now when I think I have everything set and hit run with I2V I get this error code regarding the (Upscale + Rife VFI) node:
[Errno 2] No such file or directory: 'D:\\ComfyUI_windows_portable\\ComfyUI\\input\\56597.jpg'
I'm new to upscalers, etc. do I need to do something specific with my original image to make this run?
Earlier I said I was new to Comfy, I really mean amateur and new to video. I've successfully generated Flux images and SDXL and now have successfully used a super simple I2V workflow, but yours looks far superior with much better results, just trying to work the kinks out. Thanks again!
@johninvrvam853 Perfect so far.
Your issue has nothing to do with Rife or upscaler. Seems you simply don`t loaded YOUR local start image:
1. Choose only option: "Single First Frame Image",
2. "Input First Frame Image: Load YOUR local image here.
And btw. how did you got PainterI2VAdvanced working??
@arkinson I dunno how he did it, but I did it like this:
Go to ComfyUI\custom_nodes
Type CMD in the top bar (Windows) to open the terminal and add this:
git clone https://github.com/princepainter/ComfyUI-PainterI2Vadvanced.git
https://github.com/princepainter/ComfyUI-PainterI2Vadvanced
@Setian91 Thank you for your reply. Yes, that`s the manual installation from github. There was no way to install it via the manager??? No node conflicts or error messages? What was the promblem - was the node just not displayed for installation?
@arkinson the package isn't visible in the manager no
Odd right?
@Setian91 Ok, thank you. One solution is to try different Channels in the manger.
@sb129 Hi - thank you so much for your review and feedback 🙂