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 always matches optimal Div16 WAN 2.2 resolutions
---
## Changes
- Changed upscale logic
- Changed Interpolation logic
- Changed resolution matching
- Optimized layout
- Gathered all settings
## Fixes
- none
FAQ
Comments (27)
Great work on this! It's still the best I2V workflow.
I have a bug fix for v2.3.
The Video Combine node in the Add Audio group gets frame rate directly from the frame rate input. If you turn on interpolation and add audio, the final video is 4x as long at 1/4 the frame rate. Getting the frame rate from the loaded video info fixes it.
2.2 is working great, nice work! Although I often get OOM during the interpolation with videos that are more than 5s with the resolution I use. I can easily do 10s with the same settings on 1.7.
Hello, I'm a 3060 TI 8G.
I've been using your DaSiWa Wan 2.1 Lightspeed version with great success.
384x6XX at 73FPS, around 140 seconds—excellent speed.
These past two days I've been trying to test 2.2 but can't get it working. Missing many nodes: Label, power lora loader, seed,
wanmoeksampleradvanced,
fast groups bypasser.
I've been wanting to update comfyui but haven't been able to manage it. Haha.
Hi! The easiest way would be to install https://github.com/Comfy-Org/ComfyUI-Manager
It can update comfyui and install most custom nodes.
The rest of the dependencies are in "About this version" and must be installed like written on their page.
Thank you so much every time!
I’m always getting great inspiration from your workflows.
That's the spirit! Use it to your liking 😏
There is new node Painteri2v that improves motion for low step lightx2v workflows.
"The video comparison above was conducted using the official WAN2.2 model + Light2V 4-step i2V Lora V1.0 (see the image below). The effects of using other models are open to further testing."
https://github.com/princepainter/ComfyUI-PainterI2V
Seems to be working from my test on your checkpoints. Not sure which speed up lora you used.
Waiting for them to add last frame connection to it for FLF. For now on their KJ version has the end_frame option.
Thanks for the NAG implementation, was getting annoyed putting it in there with each new workflow XD
This looks nice! Thanks for the hint. I was thinking of separating my workflow into multiple approaches (basic, FLF, audio...aio), because more and more features may overwhelm. I'll have a look!
Not messing with the new audio settings; it outputs 3sec audio for 5sec video.
That's a limitations of mmaudio, it needs 24 fps to do 5s.
What an absolute lad you are ♥ bless you
Finally an all-in-one workflow that makes easy loops in 2.2 - excellent work and ridiculously fast. 8 second clips on a 3060 in under 5 minutes.
I gotta ask, what videocard are you using?
I am able to do image to video to interpolate just fine.
I am able to do image to video to audio just fine.
It seems to crash on me when I do interpolation to audio.
All I can think is that maybe my 16 gigs of video ram is not enough?
There is no upscale of audio. Maybe you are doing something wrong. I have 16gb VRAM and all is working fine, except at some resolutions virtual memory is needed.
@darksidewalker
Sorry, interpolation.
If I turn on interpolation on it's own it works. If I turn on audio without interpolation it works.
I can't seem to get interpolation and Audio to work.
I corrected my original post.
@Russader Maybe you have too less virtual memory. But interpolation and audio won't produce synced audio. Mmaudio only works with 24fps. Anything else will cut the audio.
@darksidewalker Shame. I tried some more things and got them both working together but it extends the video back to 24 fps like you said so an interpolated 5 second video becomes 12 seconds in slow motion with 5 seconds of audio.
For those who REALLY want both it looks like it's gonna be a small pain of taking the audio generated and the good interpolated video and injecting the two together in a movie maker or something.
I will say, this is the fastest workflow I've ever used and I thank you. Seriously, thank you.
@Russader Yeah audio is not so stable. You want your video on 24fps generated without interpolation and pass this to the MMAudio.
I just removed this from my latest workkflow and added foley instead. This is not so NSFW but can handle more framerates.
Maybe I'll do an experimental WF with MMaudio.
Hi, bro!
I used your checkpoints "DaSiWa WAN 2.2 I2V 14B Lightspeed". Then when I tired to use this workflow to generate a video, I got an error message:
The shapes mat1 and mat2 cannot be multiplied (462x768 and 4096x5120)."
Could you tell me why?
Thx a lot!
You are using the wrong CLIP
@darksidewalker It worked! Thank you so much for your help.
Sorry, I only recently started using this workflow. I'm having trouble downloading it after dragging it into ComfyUI; it says there are many files, but I can't download them. How can I fix this? Thank you.
All dependencies are written inside the workflow and on the page "About this version".
The files must be downlaoded, just click the links. And the custom nodes must be installed, manually or with comfyui-mamager.
@darksidewalker FIXED thanks a lot
sorry i have an issue here, my outcome is just very blury , i noticed at the checkpoint section , you write "Choose the type of Checkpoint to use. Wan 2.2 uses a different diffusion model for the first steps (high noise) vs the last steps (low noise)." does it mean i need to use your checkpoint high and low versions? like put 2 models ?
and i set the resolution as 1980*2880, is it too low?🤣
great work btw, thank you very much for your workflow❤️
I just updates all workflows. So I'll respond to the new uploaded ones. They are easier to use.
You need both HIGH and LOW (this would be a good read: https://civitai.com/articles/20293/darksidewalkers-wan-22-14b-i2v-usage-guide-definitive-edition)
The rest, resolutions and settings are all well described inside the workflow. In the new WF there is no need to set a target resolution any more, only a start resolution.
@darksidewalker thank you very much !!!