
None
None
Long Story Video Skill
Turn your idea into 10s–10min video—AI writes scripts, prompts, and generates footage automatically.

Ads Video Skill
Generate professional ads and sales videos—AI auto-generates scripts, prompts, and footage.
3D Science Explainer Video Skill
Convert scientific concepts into stunning 3D explain animations
Feedback
freeTrialImage.bannerPity
freeTrialImage.upgradeUnlock
- ✓freeTrialImage.benefitHd
- ✓freeTrialImage.benefitWatermark
- ✓freeTrialImage.benefitUnlimited
Minimax H3 Video To Video
Minimax H3 Video To Video puts the people from your photos into a clip you have already cut, while acting, timing and soundtrack are preserved.
All Tools
Discover our comprehensive AI-powered animation toolkit
MiniMax H3
MiniMax H3 AI Video Generator
Seedance 2.5
The Future of AI Video Is Here.

Seedance 2.0
The Future of AI Video Is Here.

Seedance 2.0
The Future of AI Video Is Here.

Gemini Omni
Gemini Omni Video Generator

AI Multi-Scene Shorts Generator
Create viral AI Shorts instantly

3D Science Video
Create 3D science videos easily

AI Educational Video Generator
Generate Educational videos easily
Why Creators Reach for Minimax H3 Video To Video
Think of it as recasting rather than regenerating: your edit keeps every camera move, cut and line of audio, and only the people on screen are exchanged for the faces you supply.
- Recasting, not regenerationNothing in the original frame is rebuilt from scratch — the model edits who appears, so the look, light and motion you already approved remain as shot.
- Up to four new faces, one photo eachSupply a single clean portrait per performer, four at most, and they inherit the leading on-screen roles in left-to-right order wherever those roles appear.
- Your footage keeps the performanceGestures, timing, lighting, lens movement, edit rhythm and the existing audio track all come through the swap, so the output fits the cut you already locked.
Running Minimax H3 Video To Video in Three Steps
The workflow is short: choose footage you are already happy with, order your portraits, and let the model handle the swap.
Where MiniMax H3 Recast Performs Best
The model edits identity alone — performances, edit rhythm and the existing soundtrack come from your footage, not from a fresh render.
Four faces, mapped left to right
Four portraits is the limit. Each one claims a leading on-screen role in the order you arrange them, identity is tracked across cuts, and a one-line prompt can reorder the mapping.
Acting comes from your footage
Gesture, timing, expression, lighting and camera motion are inherited from the source, while faces and wardrobe are drawn from the photos you upload.
Edit and soundtrack survive
Your cutting rhythm and the original audio track pass through the swap, so the output can go straight back into a locked timeline with no rebuild.
Cheap previews, clean finals
Run a 768p pass to confirm the swap before paying for 1080p, and since runtime mirrors your input, cost follows clip length.
Hair and wardrobe come from the photo
A reference that shows more than a face also carries hair and clothing across, which is why a half-body shot outperforms a tight headshot.
Made for takes that already work
Studios, brand teams, social creators and fashion or commerce crews use it when a scene is perfect but the wrong performer is standing in it.
Recasting Questions, Answered
Everything creators ask before swapping performers in a clip that is already finished.
What exactly does the tool change?
It exchanges the people in an existing video for the ones in your uploaded photos. Motion, camera work, cuts and audio are kept, and by default each portrait claims a leading role from left to right across every shot that role appears in.
Is there a length limit?
5 to 30 seconds, with no single shot longer than 15 seconds, so split longer takes before uploading. The recast returns at the same length as the input.
How many people can I swap, and in what order?
Up to four, one portrait each, and they take leading on-screen roles left to right by default. A prompt like 'the woman on the right becomes person 1' overrides that order.
Which reference photos give the best result?
One sharp, front-facing, evenly lit portrait per person. A half-body frame carries hair and clothing better than a tight headshot. Skip sunglasses, hats, hands over the face, group shots, and low-resolution or heavily filtered images.
Should I preview at 768p or go straight to 1080p?
Preview at 768p to confirm the swap reads correctly, then render the 1080p master. Cost scales with clip length rather than with resolution alone.
When is this the wrong tool?
Skip it when you need new motion, new camera moves or a different environment. Results also weaken with busy crowds, small or distant faces, fast action, occlusion after hard cuts, and identity can drift toward the original performer in long shots or profile turns.
Recast Your Next Clip with Minimax H3 Video To Video
Bring your portraits into footage that is already finished and let the model handle the swap while timing, camera work, cuts and audio hold steady. Then read the Recast guide or compare the related MiniMax, Hailuo and Kling models.
