Q: Why should I download yet another WAN 2.2 I2V workflow?
A: Dozens of reasons. For one, I don't pitch runninghub or worthless videos. Only functionality I use every day and carefully maintain.
Completely overhauled I2V Swiss army knife workflow that I used for multi-segmented scenarios, POSER edition.
What the hell does that mean?
Do you wish you had:
Ability to reimagine your subject in any setting? Stand them, sit them, kneel them ... all with just a portrait?
Not 3, not 9, but dozens of perfectly rendered images of your subject in whatever pose up to 4096 x 4096 pixels or a video in that resolution? Now you can.
No need to struggle to make realistic videos that do not look like plastic? Struggle no more.
Default WAN encoder or Painter I2V. LightX2V 10/30 or 10/22. All pre-configured for you and switchable in one click. Dial the creativity of motion up or stick with the proven baseline.
I2V or FLF with just a click
Post-processing
Frame Skip
Basic Upscale or
TensorRT Upscale for perfect picture
Frame Interpolation
Multiple outputs
Pre- or post-processing results
Save last frame to keep generating new segments
Save all frames for ultimate flexibility. Each frame a picture-perfect rendering of your subject
Save original or frame interpolated video
Inputs that are hard to screw up.
Portrait or landscape? Do not guess, it will be automatically calculated.
Inputs are color coded green; nodes you might want to configure are color coded dark blue. Configurable options are in convenient control box.
Frame rates, dimensions, etc. are adjusted based on options you choose.
Less custom nodes to download. Thank you, it was a pain to clean up.
Add secret sauce to your prompt to enjoy all the capabilities of the posing LORA.
Will there be updates? Why yes, I am thinking of including a better frame interpolation and of course more and more and even more features and scenarios, but only if you watch my videos and run stuff on RunningHub for 1000 free credits. Just kidding, I don't do this crap.
Dependencies:
Courtesy of The_Cook:
https://civarchive.com/models/1648982/wan-nsfw-posing-nude?modelVersionId=2391828
LightX2V loras:
High
https://huggingface.co/Kijai/WanVideo_comfy/tree/main/LoRAs/Wan22_Lightx2v
https://huggingface.co/lightx2v/Wan2.2-Distill-Loras
Low
Custom Nodes
https://github.com/kijai/ComfyUI-WanVideoWrapper
https://github.com/chflame163/ComfyUI_LayerStyle
https://github.com/rgthree/rgthree-comfy
https://github.com/kijai/ComfyUI-KJNodes
https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite
https://github.com/yuvraj108c/ComfyUI-Upscaler-Tensorrt (a bit of pain to install but is worth it)
Description
This version was not the beginning. There are neither beginnings nor endings. But it was a beginning.
FAQ
Looks like we don't have an active mirror for this one right now.
This archive is kept alive by community contributions, so there are still some gaps, especially for newly removed content.
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.