Workflows / MiniMax H3
Premium workflow · new
Character Video Swap — one photo, any performance
A guide video gives the moves. Your photo gives the person.
Load a performance clip and one photo: MiniMax H3 re-renders the clip with that person in it — same choreography, camera, lighting and background, new face, hair, body and outfit. The guide's own audio is kept. Built on the H3 character-swap LoRA with the guide anchored frame by frame, so nothing is invented that the video did not do.
One-time purchase. Lifetime updates. Included in Pro.
You get
- Character Video Swap H3 workflow (ref2va + character-swap LoRA + guide anchoring)
- The trained body_swap prompt brief, ready to edit
- Works with registered Characters' reference photos
You need
- The Gmanski RunPod template (ComfyUI 0.42+)
- A guide clip (5–8 s, one performer) and a clear photo
- 48 GB+ VRAM
Model packs · Video · Characters · Edit an image
minimax-h3-ref2vminimax-h3-character-swapOn the template: gmanski-packs install minimax-h3-ref2v minimax-h3-character-swap
Pairs well with.
MiniMax H3newH3 Image Edit — up to 8 references
Edit a picture with other pictures, in 8 steps.
Krea 2newBulk Character Swap — Dynamic LoRA
A folder of photos in. Your character in every one — and the right LoRAs picked per picture.
Krea 2newBulk Reference Swap — no training
One reference photo. Every picture in a folder, re-cast — no LoRA needed.