kylix98/actionhug

ActionHug – LoRA for Natural Affection Motion (WAN 2.2 / I2V Compatible)

<Gallery />

Model description

ActionHug is a motion-focused LoRA designed for WAN 2.2 and I2V pipelines, adding a smooth, natural “approach and back-hug” action to any character without changing their identity or body proportions. Trained exclusively on AI-generated clips, this LoRA generalizes well across different scenes, outfits, and poses.

Trigger words

You should use actionhug to trigger the image generation.

Download model

Download them in the Files & versions tab.