<div data-youtube-video="https://youtu.be/ETAVoT2e8pM"><iframe src="https://www.youtube.com/embed/ETAVoT2e8pM" width="853" height="480" allowfullscreen></iframe></div> 
 
AI-KSK ComfyUI workflow: OpenVDN DMD8 Depth-Guided Video Gen. 
 
This workflow is the OpenVDN DMD8 depth-video-guided generation route: it takes a reference video plus depth map video, extracts the action, camera motion, and spatial structure, then regenerates a completely new subject and scene on that same motion skeleton. It runs on the MiniMax H3 hybrid FL2VA/Ref2VA INT8 model with the pinned OpenVDN DMD8 branch composed through the VDN model composer, so motion and timing are preserved while the world around them is rebuilt. 
 
Main features: 
- Depth-video-guided structure transfer: motion, camera move, and timing from the source video 
- OpenVDN vdn-minimax-h3 pinned branch with stage_dmd_8nfe via the VDN model composer 
- MiniMax H3 hybrid FL2VA/Ref2VA INT8 model with Qwen3-VL encoder 
- Two generation branches in the graph, one output node is bypassed by design - leave it off 
- SaveVideo output with audio VAE decode 
 
Suggested workflow: 
Feed a source video with clear motion and a matching depth map video (the depth extractor workflow in this series produces exactly that). Describe only the new subject and scene in the prompt; the motion and camera come from the depth-guided branch. If the subject drifts from the depth structure, lower the prompt detail rather than adding more. 
 
RunningHub workflow (try online, no install needed): 
https://www.runninghub.ai/zh-cn/post/2103500778686521345?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/BV1pPhU6XERp/ 
 
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

OpenVDN DMD8 Depth-Guided Video Gen

<div data-youtube-video="https://youtu.be/ETAVoT2e8pM"></div> 
 
AI-KSK ComfyUI workflow: OpenVDN DMD8 Depth-Guided Video Gen. 
 
This workflow is the OpenVDN DMD8 depth-video-guided generation route: it takes a reference video plus depth map video, extracts the action, camera motion, and spatial structure, then regenerates a completely new subject and scene on that same motion skeleton. It runs on the MiniMax H3 hybrid FL2VA/Ref2VA INT8 model with the pinned OpenVDN DMD8 branch composed through the VDN model composer, so motion and timing are preserved while the world around them is rebuilt. 
 
Main features: 
- Depth-video-guided structure transfer: motion, camera move, and timing from the source video 
- OpenVDN vdn-minimax-h3 pinned branch with stage_dmd_8nfe via the VDN model composer 
- MiniMax H3 hybrid FL2VA/Ref2VA INT8 model with Qwen3-VL encoder 
- Two generation branches in the graph, one output node is bypassed by design - leave it off 
- SaveVideo output with audio VAE decode 
 
Suggested workflow: 
Feed a source video with clear motion and a matching depth map video (the depth extractor workflow in this series produces exactly that). Describe only the new subject and scene in the prompt; the motion and camera come from the depth-guided branch. If the subject drifts from the depth structure, lower the prompt detail rather than adding more. 
 
RunningHub workflow (try online, no install needed): 
https://www.runninghub.ai/zh-cn/post/2103500778686521345?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/BV1pPhU6XERp/ 
 
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

Files with SHA256: 9bf1aea5c0a7c2b171cdf0e479906f42397c9a116ab15a4d0e1739d8a942b54c