Multi-Character | SteadyDancer | One-to-All | All Versions Combined
This is a fully unified Wan Animate ecosystem workflow built inside ComfyUI.
Instead of using multiple separate JSON files for different Wan versions, I merged everything into one clean, modular structure.
Included in this workflow:
Wan Animate 2.2
Wan SCAIL
Wan SteadyDancer
Wan One-to-All
Structured multi-character routing
Modular grouped node layout
Everything is organized so you can easily switch between animation styles without rebuilding pipelines.
How To Use
Upload your character image into the image input node.
Upload your reference / driving video.
Select the animation pipeline you want to use:
2.2
SCAIL
SteadyDancer
One-to-All
Important:
Enable only ONE animation section at a time.
Disable the others before generating.
Each module is clearly grouped so you can toggle easily.
Who This Is For
Advanced ComfyUI users
Multi-character animators
AI short film creators
Users tired of switching between different Wan workflow files
Guide & Updates
A full updated walkthrough guide will be posted on my YouTube channel explaining:
Proper routing
Best parameter settings
VRAM optimization
When to use SCAIL vs 2.2
Multi-character handling
Channel:
https://www.youtube.com/@VionexAI
Please wait for the guide if you are new to Wan pipelines.
Description
FAQ
Comments (16)
This looks promising add all to same, thanks! I tested wananimate and get error:
LynxInsightFaceCrop
numpy.dtype size changed, may indicate binary incompatibility. Expected 96 from C header, got 88 from PyObjec
disable the face reference image 2 in wan animate
@Vionex Yeah, it helps. Thank u! I think, error is related to numpy 2.2.6 & insightface 0.7.3 versions what i have now.
@Zepes same error, did you get to fix it?
@learnrijo no
@Vionex How can I create a dance video form image following the pose extracted from the video using Wan Animate 2.2. section? Eg the opposite of what it does by default.
I want to extract movement from video and use the image as reference to create the dance from the photo. Can this section do this or it can only insert person from photo into the location of the video?
Hey ! Thank so much. I have a question concerning Wananimate since it is probably the one working best for me.. but it has a slight problem when I try to make a longer video : the image starts burning, I guess some effect stacks every 81 frames.. any thoughts to avoid this ?
This is literally my first interaction here. If you still need help, look into flf and SVI
Hi, could you make a video tutorial? Because there are so many errors and missing nodes, even after downloading it, it doesn't work for me, I wonder where I'm going wrong.
I second the motion.
I third that motion!
i 4th this motion. I can never figure this crap out and everyone just says use the comfi ui manager, witch is as helpful as a bag of bricks.
And now I figured it all out and its a top notch wf. Super cool dances and realistic moves with SCAIL. Best section out of 4. Nothing comes close to the quality of the motion transfer. I even added a Uni3C camera motion path, to also capture camera movement. Just keep at it, it will suddenly click.
WanVideoSampler
Sizes of tensors must match except in dimension 0. Expected size 22 but got size 8 for tensor number 1 in the list.
how should this be?
Hope you upload that video guide soon, this looks promising.
Wan-OneToALLAnimation has wrong GET node for ExtendVideoDuration and Video Combine 🎥🅥🅗🅢
Get_vae, Get_pose_images, and Get_num_frames should be Get_vae_0, Get_pose_images_0 and Get_num_frames_0