My ComfyUI workflows for using Wan 2.2
This workflows are used by me to create my art.
They are optimized for my checkpoints and created of my latest knowledge to enhance the outcome.
"If this workflow leveled up your day, I'd purr-eciate a like! 😻"
Versions & Information👇👇👇👇👇👇👇👇
👉 Please read below and the file descriptions "About this version" for more info's.
🎬🐈Click me! HowTo + Video
💡Some WAN 2.2 versions use high+low checkpoints, other like S2V use only a single checkpoint - make sure to read the descriptions and use the correct checkpoints.
⚠️ Do not use the workflows with the "Nodes 2.0 beta" from ComfyUi or it will mess up things.
👇👇👇👇👇👇👇👇
What you get from the comfy workflows:
♨️ Easy controls
✅ As less as possible dependencies
🪧 Detailed documentation
⛓️ Highly automatic logic
✨ Optimized results
🎬 Fully automated resolution logic
🔖 Bookmark-Shortcuts with number keys
Types of workflows
FastFidelity C-AiO
🖼️ I2V and FLF2V
🧩 Automatic aspect‑ratio calculation and fitting
✨ Multiple Upscalers
Torchlanc (very fast, color correct, low VRAM)
Upscale with Model (additional detail, high quality)
RTX Super Resolution (ultra fast, very accurate)
🤝 Video resolution matching - Fully automatic scaling and resolution calculations
🕠 Length automation - Fully automatic calculation of frame count
🫥 Watermark option
🧮 Color match feature
👾 MiniMeme feature - Create small gif's
🎭 NAG - Negative prompting with CFG1
🪄 Interpolation feature
🌀 Perfect loop feature
🃏 Last Frame Extraction
🔖 Bookmark-Shortcuts - with number keys
FastFidelity C-SVI
🖼️ SVI (SVI 2.0 up to 10 samplers)
🧩 Automatic aspect‑ratio calculation and fitting
✨ Ultra‑fast, color‑correct upscaling (torchlanc) OR Upscale with Model
🤝 Video resolution matching - Fully automatic scaling and resolution calculations
🕠 Length automation - Fully automatic calculation of frame count
🫥 Watermark support
🪄 Interpolation feature
🔖 Bookmark-Shortcuts - with number keys
Swarm Basic
Absolute basic SwarmUI preset
Not recommended over ComfyUI for video generation
Backend Test
If this does not create the example video your ComfyUI backend is broken.
If this works, but your other workflow not, the other workflow is broken or missing dependencies
🩻 Known issues and advice's
⚠️ Some workflows may set on webp av1 encoding (VHS node) - If your computer/setup missing drivers use any other like H265 or H264!
Install ffmpeg!
Update Comfyui and custom_nodes!
Update pytorch 2.9+cu128 or higher
Make sure to read where files/models should be placed inside the workflow
Check if the filepath for model/clip/vae match your system like Linux/Windows
The plugin ComfyUI-DD-Translation can break node connection (avoid)
All older Versions are available inside my GitHub Repo.
Spacial thanks to @Abyss_Games for a really good idea and help with the "loop"!
I got some really good ideas from @Gladas workflows!
YOU are responsible for outputs as always! If you make ToS violating content and I get aware I WILL report this.
Description
## Requirements
- https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite
- https://github.com/rgthree/rgthree-comfy
- https://github.com/Artificial-Sweetener/comfyui-WhiteRabbit
packaging, torchlanc
- https://github.com/kijai/ComfyUI-KJNodes
- https://github.com/yolain/ComfyUI-Easy-Use
---
## Features
- 🧩 Automatic aspect‑ratio fitting
- 🖼️ Monitor‑resolution selective upscaling
- ✨ Ultra‑fast, color‑correct upscaling (torchlanc)
- 🔖 Bookmark-Shortcuts - with number keys
- 🤝 Video resolution matching - always optimal Div16 WAN 2.2 resolutions
- 🫥 Watermark support
- Separated upscale logic for VRAM saving
---
## Changes
- none
## Fixes
- FLF not working (missing image)
FAQ
Comments (37)
Where is the settings for Sampler/Scheduler? Or is it already ingrained in the workflow? Using ComfyUI
It is pre-set. You can change anything to your liking inside the nodes and subgraphs.
I have a problem. I'm a kettle, I don't understand almost anything about these nodes, I downloaded Workflow, I seem to have done everything, inserted the original image, but in the animated version I just have 80% of the video covered with blue noise, through which only 20% of what is happening in the video is visible.
Did I do the right thing by downloading your checkpoint and downloading only it, or did I need to install something else?
I also want to know if something needs to be reconnected or connected as soon as I open this WorkFlow? If possible, give me the exact name of the nodes that need to be connected to each other.
You need the correct VAE and CLIP (links on main page) and install all dependencies of the custom nodes.
Rewire is not necessary.
@darksidewalker I have everything installed, I am 100% sure that the nodes, VAE and CLIP are installed the right ones.
@RussianMan_BRUTAL Do you mixed high/low checkpoints or set one twice?
@darksidewalker This is a good question, there are two identical "Upload checkpoint" nodes in Workflow, I installed a model in each of them, which I downloaded from the same author "dasiwaWAN22I2V14B_radiantcrushHigh", so it turns out that this model alone is currently in two nodes at once. Only one of these models is listed in the links on the main page. What am I doing wrong?
@darksidewalker I uploaded the same model to two control points, This one:
https://civitai.com/models/1981116/dasiwa-wan-22-i2v-14b-lightspeed
What's wrong?
@darksidewalker Are you saying that I just needed to download the second version of the MidNightFlirt low model? I found out a second ago that it turns out that you can choose the model version on civitai
@RussianMan_BRUTAL If you are completely new to WAN 2.2, I suggest reading that: https://civitai.com/articles/20293/darksidewalkers-wan-22-14b-i2v-usage-guide-definitive-edition
You need both HIGH and LOW checkpoints.
I'm getting this error with the new 1.6 workflow. I'm basicaly using it as is with one high and one low lora. tried I2v and FTLF , both give the same error: RuntimeError: Given groups=1, weight of size [5120, 36, 1, 2, 2], expected input[1, 64, 29, 78, 60] to have 36 channels, but got 64 channels instead
The error pops up at the begining of the process for ksampler high
Wrong VAE used.
@darksidewalker -_- thx lol
@darksidewalker I also had the same situation, I am using vae correctly, KSamplerAdvanced has an error display:
Given groups=1, weight of size [5120, 36, 1, 2, 2], expected input[1, 32, 21, 72, 48] to have 36 channels, but got 32 channels instead
I haven't been able to find a solution, please help meeeeee!!!!!!
@MuYu_Ender No, the error comes from the wrong VAE, so you are not using it correctly. 😉
@darksidewalkerThe VAE I tried was wan2.1, that is, downloaded from the link you gave, and used the beta workflow today to try to get an error.
I only modified the selection of the Checkpoint, using your Checkpoint：DaSiWa WAN 2.2 I2V 14B Lightspeed v7，A "high" and a "low", then add a umt5_xxl (also downloaded from the link you gave), and other than that, no parameters are moved, and finally the image is loaded and the KSamplerAdvanced error I mentioned above is obtained
As a novice, I couldn't find a solution, and I asked about AI, but to no avail,Can you tell me where the problem went wrong? I don't know which step I did wrong .
@MuYu_Ender I do not know what to say, the error appears if you select the wrong vae. There is no other know dependency I know that can cause this error.
how do you make it loop? i made a 89 frame video, and it saved the end image. i can load that then into new starter image and make a new vid...but its two seperate videos....i made my own flow that uses the last frame to use as a load image of next video....so its 6 img to videos on same flow as running into each other, and then i take all the outputted vids and auto stitch em into one.
im jut looking for a better setup than i got and i thought this would do that.
You are talking of separate thing here.
Do you want to make a loop or do you want to extend a video with multiple videos?
@darksidewalker I would see extend a video with multiple videos workflow :-)
@darksidewalker not to hijack the conversation, but how would i go about making video a full loop? as in idle animation of a character with 81 frames that seamlessly returns to first frame at the end? i am completely new to comfy, so not sure which nodes i have to activate for that to work :)
@MarkMix You can do it with my workflow and just use the loop-feature with the same start and end image.
@darksidewalker thanks, i figured it out after turning off beta 2.0 nodes, feature option was hidden under another window for me
@MarkMix Oh yeah the beta 2.0 nodes messing all up ... maybe I should say this on the page.
In workflow FastFidelity C-AiO v1.6 always the same errors:
No input found for flattened id [569] slot [-1] when i try to run I2V.
No input found for flattened id [543] slot [-1] when FLF2V.
Works only if i delete the whole FLF2V section.
These nodes are the input images on my side, maybe you did not set one?
@darksidewalker but all that nodes with errors are bypassed. Why they are calling errors? I didn't even change anything, just open workflow, upload image for the i2v and errors appear after clicking Run button. May i ask you to upload fixed version?
Bypassed nodes do not produce error normally. They are not processed. You should check your comfy installation and if all dependencies for custom nodes are installed.
@darksidewalker oh, i think problem was because of older version of my comfy. After an update no errors detected. Sorry for bother
I have the same issue with this workflow. Nightly comfy version installed. No input found for flattened id [558] slot [-1]. Won't let me do anything with the workflow, even when I tried to export the workflow to save my settings, same error. There are no visible individual nodes erroring out.
Scratch that, manager was outdated, so not pulling latest version. Updated comfy manager, which updated comfy to 0.5.1...now basically everything is broken. Can't even select the model in the checkpoint node, because it's broken and brings up a blank field.
@Minase460 So you basically destroyed your comfy install. I would recommend a fresh install.
Scratch yet again.... updating comfy deleted my mklinks for some reason, resulting in no results when selecting checkpoint. Updated all custom nodes. Rebuilt my mklinks. Workflow working now generated my first vid 65 frames in 49.9 seconds. Working as intended, and very fast. Just posting here for people with same issue.
So maybe this is a dumb question, but in prior workflows I could see what was going on in the sampler with previews enabled, and kill it if the gen was bad and reprompt. Is there a good way to get a preview on this workflow? I haven't figured out how with the samplers buried in a subgraph
If i get ur idea correctly then just create or duplicate video combine node and connect image input with Any Swith (rgthree) image output
I'll soon release a workflow without the subgraphs and some other optimizations. As they are a bit buggy atm. You may just wait or you have to expand the current graphs.
Looks like we don't have an active mirror for this file right now.
CivArchive is a community-maintained index — we catalog mirrors that volunteers upload to HuggingFace, torrents, and other public hosts. Looks like no one has uploaded a copy of this file yet.
Some files do get recovered over time through contributions. If you're looking for this one, feel free to ask in Discord, or help preserve it if you have a copy.