MiniMax H3 8 Step Wushu Motion Control

AI-KSK ComfyUI workflow: MiniMax H3 8 Step Wushu Motion Control.

This workflow is the 8-step wushu motion control final version. It generates martial arts action video with the MiniMax H3 hybrid FL2VA/Ref2VA bf16 model, the LightX2V 8-step turbo LoRA, and a dedicated wushu LoRA stack (wushu_e3 FL2VA, spatial physics, Motion Repair), tuned for believable fighting motion and impact timing.

Main features:
- LightX2V turbo 8-step LoRA (768p v1.0 ComfyUI bf16 build) on the model path
- MiniMax H3 hybrid FL2VA/Ref2VA bf16 model with Qwen3-VL 32B encoder
- Wushu LoRA stack: wushu_e3 FL2VA 1000-step (0.75), Motion Repair (0.5), spatial physics (0.3)
- Connected 10-step simple schedule with euler sampling
- Video and audio VAE decode to H.264 MP4 with audio at 24 fps

Suggested workflow:
Use this route for martial arts fights, weapon moves, and fast impact shots. Give one clear action per sentence (strike, dodge, follow-through), keep the two combatants framed with enough headroom, and check contact timing on a short clip before rendering the full take. One branch in the graph is bypassed by design; leave it as is.

RunningHub workflow (try online, no install needed):
https://www.runninghub.ai/zh-cn/post/2100146680422281217?inviteCode=rh-v1111

AI-KSK RunningHub workflow hub:
https://www.runninghub.ai/user-center/1892566146468511746?inviteCode=rh-v1111

Bilibili tutorial:
https://www.bilibili.com/video/BV1MYec6uEoN/

RunningHub registration and AI tools:
https://www.runninghub.ai/?inviteCode=rh-v1111
https://www.runninghub.ai/ai-generator/2046514150500524034?inviteCode=rh-v1111
https://www.runninghub.ai/vip-rights/3?inviteCode=rh-v1111
https://rhtv.runninghub.ai/projects?inviteCode=rh-v1111

Stable global access:
https://lelians.com/#/register?code=mFIJ5tZn

TokenRhythm AI model credits:
https://tokenrhythm.studio/i/rf_tr_kM6wekDquDDZQD4PUNmgiuHA

Youyun Cloud GPU:
https://passport.compshare.cn/register?referral_code=A7fTQUYY9lCDFJwA6MMvoq
https://www.notion.so/2b72459a3ac6808f8282f2c687ed8f52

RunPod Cloud GPU:
https://runpod.io/?ref=sxk5n5m7

Support AI-KSK:
https://ko-fi.com/aiksk