Skip to the creator
AI VIDEO · YOUR SCENE

I Love The SmithTry the scene from either side.

Who notices the music first? Put one person behind the headphones and the other in charge of breaking the silence. Swap the portraits and see which casting feels more like you.

Clear left and right rolesSwap photos in one clickSample portraits to explore
Your creative workspace

Choose who breaks the silence

01Listener on the left, speaker on the right

Add one adult portrait for each role. Swap roles reverses the two uploaded references. You can explore the layout with the clearly marked source sample portraits.

Start with

The trend supplies the direction. Text mode uses your own description; first-frame mode animates one image; references guide a new scene.

02Your role-based scene
03Choose your output
MiniMax H3768P · 15s · Auto
24 credits

MiniMax H3 · 480P / 768P / 1080P: 4–30 whole seconds · 2K: 4–15 seconds. Trial and video-pack units cover only five-second outputs at their matching quality, without a video reference. Other settings use general credits.

Add a creative detailOptional

Credits follow duration and quality. Outputs over 15 seconds double the full video fee. View credit plans

Reference exampleVIDEO

The selected preset guides the new scene. The example and your render can differ in timing, sound and appearance.

Check the casting before Create

Swap roles reverses your references; existing videos are unchanged.

01 · Left · headphone listener02 · Right · starts the conversation
A little inspiration

Two portraits, one encounter

The reference video and sample portraits come from the source site. They illustrate the assigned roles and are not a generated result from this service.

CASTING MAKES THE JOKE

The same two faces can tell a different story.

The listener’s pause and the speaker’s confidence give the scene its contrast. Assign the roles deliberately, then use a small reaction rather than a complicated action sequence. Swapping the reference photos changes the requested casting; it does not edit an already generated video.

01

Left: the headphones make a private moment visible.

02

Right: a short remark becomes the invitation to respond.

03

Together: the shared look resolves the encounter.

Make it your own
How it works

Try the casting before you commit

Exploring roles and samples does not create a video. Generation is a separate action.

01

Add the two faces

Keep one adult per photo. Use similar crop sizes so neither role starts at a disadvantage.

02

Try the reverse

Press Swap roles to reverse the references. Check the labels before choosing your output settings.

03

Create your preferred version

Confirm your permission, inspect the quote and generate. Review the finished clip in your results.

A better starting point

Give each role a clear starting point

Keep the eyes visible

Sunglasses, hair across the face and heavy shadows weaken expression cues.

Use similar face sizes

A tiny face beside a tight close-up can make the pairing harder to interpret.

Swap before generating

The role swap updates the input photos. A new casting needs a new generation request.

Know the cost first

Explore first. Review the quote next.

MiniMax H3 starts at 15 seconds and 768P for this scene. Change duration or quality to update the general-credit quote, and review it before submitting.

MiniMax H3 · 480P / 768P / 1080P: 4–30 whole seconds · 2K: 4–15 seconds.

Default scene duration · general credits per video
ResolutionDurationcredits
480P15s16
768P15s24
1080P15s38
2K15s40

The quote updates with duration and quality. For 16–30 seconds, the whole video fee doubles, including any video-reference reserve. Up to five photos and audio add no extra fee. Standard queue is used.

Trial and video-pack units cover only five-second outputs at their matching quality, without a video reference. Other settings use general credits.

View credit plans
A few things to know

Before you make the scene

Do we have to be a couple?

No. Two consenting adults can play the roles as friends, partners or colleagues. The prompt uses listener and speaker rather than assigning a gender.

What does Swap roles change?

It reverses the two input references. The left role remains the headphone listener and the right role remains the speaker.

Can I try it without my own portraits?

You can load the marked source samples to inspect the arrangement. Generating with any images is still a separate action with a displayed quote.

Will swapping edit my finished video?

No. It only changes the next request’s input arrangement. Existing results remain in your creations; generating a second version has its own cost.

How much does a render cost?

The editor shows the HiHD credit quote for your selected model and output settings before you generate. The Pricing page lists the available plans and credit packs.

I Love The Smith

Let the other one say it first.

Swap the roles before you choose your take.

Start creating