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
face swap minimax h3
Replace the cast of a finished clip with face swap MiniMax H3: four photos in, and the timing, camera work and audio you recorded come out the same.
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.
Veo3.1
Create Stunning Videos with Veo3.1
FLUX 3 Video Generator
Kling 3.0
Next-Gen AI Video Generator
Grok Video Generator
Create Videos from Text or Images with AI
MiniMax H3 video generator
MiniMax H3 AI Video Generator
What face swap MiniMax H3 Changes — and What It Keeps
Recasting starts from footage that already works: the take sets the acting, the timing and the sound, while your photos decide whose face appears on screen. Nothing is re-shot or re-generated.
- One Photo per Performer, Four at MostGive the tool a single clear picture for each new face. Those faces take over the leading figures in every shot they appear in, matched left to right — so the order you upload in has to mirror the order on screen.
- Acting, Lighting and Lens Moves Stay PutNothing fresh gets animated: gestures, beats, expressions, scene light and the way the camera travels all come from the original footage. Only faces are replaced — and the wardrobe, when your reference shows it.
- Cheap 768p Preview, Then the 1080p FinishA quick low-resolution pass confirms the casting before you commit to a 1080p final, and the bill follows clip length instead of a flat fee.
How to Run a face swap MiniMax H3 Recast
Three moves — load the source, line up the photos, then preview and render. Length and audio come out exactly as they went in.
Where face swap MiniMax H3 Performs Best
Picture a casting desk laid over footage you have already shot. No fresh render is requested — the take stays exactly as filmed, and you simply choose which performer's face occupies each role.
Prompt Control Over Who Becomes Whom
An optional short prompt overrides the default left-to-right mapping — say, telling the tool that the woman on the right should become person 1.
Faces Tracked Through Every Cut
A swapped face stays consistent across the edit, because identity is followed from shot to shot wherever that person appears.
Edit Rhythm and Audio Left Alone
Cut points and the soundtrack pass through untouched, so a recast slots into a finished clip instead of being rebuilt from zero.
Runtime Identical to the Source
The result keeps the duration of whatever you uploaded, which means no re-timing or re-editing, and the price scales with clip length.
What a Strong Reference Photo Looks Like
Sharp, front-facing and evenly lit, with one person per image. A half-body frame preserves hair and clothing better than a tight headshot, while sunglasses, hats, covered faces, group photos and heavy filters all weaken the result.
Where It Struggles: Crowds, Distance, Fast Motion
Dense crowds, small or far-away faces, rapid movement and occlusion right after a hard cut are the weak spots, and identity can drift back toward the original actor in long shots or profile turns.
face swap MiniMax H3 — Common Questions
Answers about recasting footage you already have with face swap MiniMax H3, from clip limits and photo tips to pricing and known weak spots.
What exactly does face swap MiniMax H3 change?
It replaces the people in footage you already own: your reference photos stand in for the on-screen cast, while motion, camera behaviour, edit points and audio stay as recorded. Nothing is built from zero.
How long can the source clip be?
Anything from five to thirty seconds. Each individual shot must stay under fifteen seconds, so break up longer takes before uploading; the output matches the length you sent.
How many faces can I swap, and in which order?
Four at most, one photo each. The default reading goes left to right, matching the main figures in frame; a brief prompt lets you reassign them when that order is wrong.
Which reference photos give the best results?
Choose a crisp, well-lit portrait with the subject facing camera, framed around the torso rather than the head alone so hair and clothes are visible. Avoid sunglasses, headwear, a hand over the face, several people in one frame, heavy filters or low resolution.
Is a 768p preview worth it before 1080p?
Yes — the low-resolution pass shows whether the right face landed on the right person, and only then do you render at full 1080p. Since pricing tracks runtime, catching a bad photo early costs almost nothing.
When is this the wrong tool?
Skip it if you need brand-new movement, a different camera angle or another location, since the uploaded clip dictates all three. It also falters on packed scenes, tiny or distant faces, and fast motion or objects crossing a face just after a cut.
Recast a Finished Take With face swap MiniMax H3
Swap the performers, not the production — the cut, the camera work and the soundtrack you already have all come through the other side unchanged.