MiniMax H3 OpenVDN DMD8 Ref2VA

AI-KSK ComfyUI workflow: MiniMax H3 OpenVDN DMD8 Ref2VA.

This workflow is the OpenVDN DMD8 reference-to-video route. It composes the pinned OpenVDN vdn-minimax-h3 model with the DMD8 NFE stage via the MiniMaxH3VDNModelComposer node on the Ref2VA branch, then decodes with the MiniMax H3 audio-video decoder and saves the result.

Main features:
- OpenVDN vdn-minimax-h3 pinned model with stage_dmd_8nfe via the VDN model composer
- MiniMax H3 Ref2VA INT8 base model with Qwen3-VL 32B encoder
- Reference image plus text prompt input
- VDN execution plan on the Ref2VA branch
- MiniMax H3 AV decode to saved MP4 video

Suggested workflow:
Use this route to compare OpenVDN DMD8 reference-to-video output against LightX2V and PDD on the same reference image. Keep the reference face stable and prompt simple on the first run.

RunningHub workflow (try online, no install needed):
https://www.runninghub.ai/zh-cn/post/2098324169208721410?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/BV16DYL6FEsh/

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