(Walkflow update) Wan 2.2 (Native) (DetailEnhancer) (Long Length Loops) Ver 20251231 (For Patreon Supporters) (Patreon)

So a lots of AI videos what do we do after generate?

As we have try out in some tutorials before AI video like Wan 2.2 can be detailer.

Previous Post:

Wan 2.2 Animate V2 Update And Video Detail Enhancer

https://www.patreon.com/posts/wan-2-2-animate-140495285

FlashVSR UpRes With 1 Step - Explore This AI Video Enhancer With Other Methods in ComfyUI!https://www.patreon.com/posts/flashvsr-upres-1-141632921

We add detail with 2nd or 3rd sampling. (or called 2nd pass for animator).

This workflow update based on the Video Detail Enhancer

Included some updated Lora setting, and for long length video (not Context Windows or 1 big Sampling).

This is based on Wan 2.2 Text2Video model.

Lora that I use (you can adjust according to your need for detailer) :

Lightx2v Wan2.2-Distill-Loras:

https://huggingface.co/lightx2v/Wan2.2-Distill-Loras/tree/main

rCM Lora:

https://huggingface.co/Kijai/WanVideo_comfy/tree/main/LoRAs/rCM

DetailEnhancerV1

https://huggingface.co/vrgamedevgirl84/Wan14BT2VFusioniX/blob/main/OtherLoRa's/DetailEnhancerV1.safetensors

Skin-Detail Lora (Thanks to Kayte in our group @kayte0342 for Lora Training)

https://huggingface.co/kayte0342/wan2.2_loras/resolve/main/skin_test_25_low_noise.safetensors?download=true

The demo video was using SVI 2.0 Pro to generated, the tutorial video post I just posted before this post.

https://www.patreon.com/posts/wan2-2-svi-v2-0-146997189

720P Output

After detail enhance, don't forget to upscale!

Have Fun :)