WAN 2.2 💠 T2V / I2V / S2V / T2I🔹4-6 steps🔹Loop 🔹SVI Video extend 🔹WanVideoWrapper Workflow🔹K3NK
NOTICE: IF YOU ARE NEW TO COMFYUI, I DONT RECOMMEND STARTING WITH THIS WORKFLOW, IS AN ALL IN ONE WORKFLOW YOU WILL GET CONFUSED.
I UPDATED v2.3, YOU CAN CHOOSE IF YOU WANT RES4LYF OR WVW SIGMA SPLITTING
Since v1.8 im using subgraphs, you need latest comfyui with the latest frontend updates, also probably you will need to update Comfyui Manager.
For me was: setting nightly vers on the manager then update confy from the manager too.
* i allways leave the image overlay enabled, it will error if you dont have an image on it, sorry :S
v2.0
You can choose whether to merge with the loras or not, change the lora connection from the model loader to the set loras node, after that uncheck merge_loras
The toggled i added for radial sage, requieres Sage Attention to be installed, if you dont have it you should unplug the conexion i made to the attention type on the model loaders..
Be sure to update WanVideoWrapper nodes ..
The workflow works just fine with Wan2.1 if you bypass first sampler and model loader
This workflow is included with all my uploads as metadata, drag and drop a vid to comfyui.
I run a 4090 locally
You can use it as other modes if you activate this option for enabling the required nodes.
Make sure you load the correct models on the model loaders
Kijai's WanVideoWrapper Node
Wan2.2 Kijai's models
Lightx2v LoRa needed for lower steps (faster generations)
Wan22-Lightning Lightx2v alternative, newer and better in some scenarios
NSFW Clip Vision (not really needed in wan2.2)
RES4LYF Nodes
MMAUDIO
https://huggingface.co/Kijai/MMAudio_safetensors/tree/main
once installed replace nodes.py file in the mmaudio node folder with this one: https://github.com/phazei/ComfyUI-MMAudio/blob/master/nodes.py
WAN 2.2 💠 T2V / I2V
MMAUDIO NSFW
https://huggingface.co/phazei/NSFW_MMaudio (download and rename it to "mmaudio_nsfw_large_44k_v2_fp16.safetensors")
BOYONODES
K3NK IMAGE GRAB
Description
Fixed the set/get nodes bug when bypassed, now they get muted instead
Added lora/model loaders for each section Extend Video
Added S2V capabilites addind kijai's workflow as a subgraph
Replaced some nodes with already existing ones for les custom_node usage
FAQ
Comments (27)
everytime I open a new workflow I get about 10 new random obscure nodes that need to be installed . there is no standardization and every workflow is a huge mess of new and old nodes. I'm getting loop detected or an infinite loop due to the mess in there.
i tried to reduce the number of nodes in the last version, but some stuff is only available in certain nodes.. if you find any nodes that ca be replaced, tell me please
My monitor won't turn on, I can't see anything, this workflow is the worst. My mouse scroll wheel is going in the wrong direction where is the mouse node? Can you help me install vram I think I put it in upside down. Why do you keep putting in crap I don't understand I want it to work now. Fix it now.
He provides workflows. Not brains.
I get a weird error when I run this (unlike your previous versions):
WanVideoModelLoader
'NoneType' object has no attribute 'lower'
updated wanvideowrapper?
Thanks for the response. Turned out it was due to the incorrect Lora model not being loaded. I am on latest wan video wrapper, and I get the following two errors when loading the latest (even after a manual install via git in windows cmd):
WanVideoAddAudioEmbedsin subgraph 'New Subgraph'
NormalizeAudioLoudnessin subgraph 'New Subgraph'
@fancypantzzz i added S2V as a subgraph, i guess i forgot to change the name, you should update confy to latest vers, like i added in the description
Cant make 1.8 working. Getting an error "Requires ComfyUI 0.3.53:
70e14245-3789-4aa6-a982-7aa2f0ab0872"
The problem is with node to the left on which is written "Open subgraph here"
New install on 0.3.55 didn't help either. Any ideas?
i think you need to have the latest frontend version, you can update it by switching to nightly in the manager, this is what i did: Switch to nightly in the Manager > Restart ComfyUI > Update ComfyUI button in the Manager.
mine is ComfyUI 0.3.55, ComfyUI_frontend v1.25.11
also inside the sbgraph theres the s2v dev branch nodes, i will upload 1.9 today with the latest nodes and a new method for video extend
on 1.8 i keep getting an error when trying to extend with new loras and prompt extension. Using 4080 16gb + 64gb ram.
Error during sampling: CUDA error: out of memory CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1 Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
it might be that i left 24 blocks to swap, if you dont have 128gb ram, it will oom the ram.. in the1.9 i reduced default settings trying to avoid you guys to get errors on your first run..
It keep proposing Loop detected: Loop (1977,2599,2005) - not submitting workflow
God damn I give up on trying to make this work, good luck to anyone with a higher level of patience
i know , sometimes simple things makes you happier instead lol
Wan2.1 is so much easier and gives me better motion with the speed loras. Gonna stick to it til wan2.2's kinks are ironed out.
it will be easier for all of us if you make a video tutorial of how to use this workflow . :P
Has anyone found all the nodes that should be used in this workflow? Is there anywhere like a list I can see?
mostly are available to download from the comfyui manager, the S2V one needs a special branch, i will upload 1.9 now tell me if you have any issues
@K3NK I can't find 24 nodes. "RAMCleanup", "VRAMCleanup" "ImageListToImageBatch", "ttN text" etc there's a lot. Any idea where these come from?
same, a lot of nodes manager can't download
I just see list of missing nodes, press install, its required comfy ui to restart. After restart i see the same list.
@legolas8456 thats strange, maybe you dont have comfy updated so those nodes wont show up, make sure you update the comfyui manager too..
@K3NK Yes, I did all the updates. I encountered a similar situation in another workflow and performed a manual installation in a convenient interface. But in your workflow, I have this situation with 15 nodes and more. I agree that it's strange... Maybe it's difficult, but I don't understand why the authors don't just pin all the nodes they use, that would solve all the problems.
comfyui-reactor-node has been disabled by github. any workaround?
@cs20105853 as far as i know, theres a non sfw version arround, google it
I don't understand how you set the version to "nightly". I see the Channel options but it's only things like latest/recent/dev/forked/tutorial, no "nightly" option exists anywhere. Can you explain this in more depth? I can't install all the nodes, which I assume is related to this