xeperWAN2.2workflow_4steps_I2V-FLF-T2V

There are far more WAN 2.2 workflows than I realized, but it's too late now :) Maybe someone will find it useful.

RTX 3060 12gb | Ryzen 5 3600 | 32gb RAM

(With models loaded, 5 second vid)

I2V & T2V: Around 4 1/2 minutes

RIFE: Adds around 30 seconds to generation time, but better FPS!

I set the res to 480 x 832, but you can change it using Resolution Master's presets. I considered adding an upscaler, but I couldn't notice a difference and it added an extra minute.

Downloads:

---

I don't recommend this WF if you don't want to download both the I2V and T2V models. If you want to try it still for some reason, add dummy files in your WAN checkpoint folder and use them as the undownloaded models to prevent an error.

Base models & VAE: https://huggingface.co/QuantStack/Wan2.2-T2V-A14B-GGUF/tree/main & https://huggingface.co/QuantStack/Wan2.2-I2V-A14B-GGUF/tree/main

I use Q3, Q4 seems to be too big. You only need one VAE from either link.

Lightx2v: https://huggingface.co/lightx2v/Wan2.1-I2V-14B-480P-StepDistill-CfgDistill-Lightx2v/tree/main/loras

CLIP (UMT5): Download with ComfyUI Manager!

---

I suspect the memory problems I've had is cos I need more RAM or something. If you have memory problems with similar specs to mine, a recent ComfyUI update was meant to fix WAN memory leaks. You can also try getting Comfy 0.3.43, then use Update.bat (not python dependencies). Pytorch 2.8 might cause memory problems, though I don't think there's been a confirmation.

For I2V, you can use Florence and leave your prompt empty unless you want something specific.

To show video previews, turn on "Display animated previews when sampling" in VHS' settings.

If you want FLF for anything other than perfect loops, use the different image switch.

The NSFW switch was made to work with CubeyAI's NSFW lora.

Delete all the filename stuff near VAE Decode if you don't want them.