Disappear in a video with AI

Your phone clip goes to three Wan 3.0 nodes that make you disappear three ways: the clothes collapse on the floor, the body turns to dust, you reappear one step away.

Interactive demo

Real workflow, real results. Pan and zoom freely.

Use this workflow

The real workflow: the starting clip wired as a reference to three Wan 3.0 Video nodes, one disappearance per node. The three videos are the demo's own, the same person and the same room.

At a glance

What the workflow produces: 3 videos of 15s in 720p, 9:16
Canvas structure: 8 nodes wired by 6 connections
Models wired in: Wan 3.0
Cost of one full run: about 180 credits, roughly $2.16

It is the cheapest hook there is: one shot, three seconds, and the viewer has already stopped scrolling. You are standing still, and your body vanishes into thin air. The empty clothes fall to the floor in a heap.

The three disappearances in the demo are nothing alike: the clothes collapsing all at once while keeping their colour and their fabric, the body dissolving from the feet up into particles carried off by a draught, and the dry vanishing followed by a reappearance one large step away.

Two conditions in your starting clip do everything: the camera does not move at all, and nothing else moves in the frame. The three prompts repeat it, because a moving background gives a dissolve instead of a disappearance. Wan 3.0 takes your clip as a reference and keeps your face right up to the instant you go. The three videos cost one hundred and eighty credits.

How it works

1
Wedge the camera
On a table, against a wall, or on a tripod. A handheld shot gives a soft dissolve: a fixed camera is what makes the disappearance sharp.
2
Stand in a wide shot
Standing, still, with floor visible under your feet. That is where the clothes are going to fall, and if there is no floor in the frame they fall out of shot.
3
Empty the background
No television on, no curtain moving, nobody walking past. Whatever moves behind you stops the model from understanding what is meant to disappear.
4
Generate all three, keep the hook
"Generate all" outputs the three disappearances. The clothes one reads best in three seconds, the dust one is the prettiest, the return one is the funniest.

Variations and use cases

The same canvas, tuned for different needs: each variation is two or three lines of prompt away.

The clothes that fall

The most readable disappearance: the body goes, the clothes collapse on the floor in a heap. It is the one that reads without explanation, even cut down to three seconds.

The dissolve into dust

The body comes apart from the feet up into glowing particles carried off by a draught, the face going last. It is the most beautiful of the three, and the slowest.

The one step teleport

You vanish, the frame stays empty for half a second, and you reappear one large step away looking at the spot you have just left. It is the comic version.

The hook of a longer video

Three seconds of disappearance open a video better than a sentence. Put them at the top, then run on into your content with the Edit node, which assembles in the browser.

The transition between two shots

Disappear at the end of one shot and reappear at the start of the next, in another place or another outfit. The cut becomes an effect instead of something endured.

Making something else disappear

The principle does not assume a person: a car, a piece of furniture, a full plate. To erase an object from a photo rather than a video, remove an object is the sheet for that.

Common mistakes

Holding the camera in your hand

The slightest frame movement during the disappearance turns the effect into a dissolve, and the heap of clothes lands crooked. Put the phone down, it is the only real constraint.

Framing too tight

In a chest shot there is no floor in the picture, so nothing to receive the clothes. Step back until you can see your feet and a little space around you.

Leaving something moving behind you

A screen on, a fan, someone crossing: the model no longer knows what is meant to go and renders a mushy picture. A perfectly still background doubles the quality of the result.

Moving during the five seconds

The stiller you are, the sharper the disappearance. A small slow gesture, like spreading your arms, works very well; walking or turning around does not.

Recommended models

Frequently asked questions

Do I need a green screen?

No, and that is the whole point: an ordinary living room is enough, and that is what the demo shows. The model does not cut your silhouette out, it replays the scene from your clip.

Do the clothes fall believably?

On the demo render, yes: they lose their shape at once and collapse in a heap, keeping their colours and their fabric. It is the variant we probed first, precisely because it is the hardest.

Can I disappear outdoors?

Yes, provided nothing moves behind you, which is harder outdoors. A courtyard, a car park, a path with no traffic suit better than a busy street.

How much does one disappearance cost?

One hundred and eighty credits for the three versions, sixty per five second video in 720p with sound. Rerunning only the one you keep costs sixty credits.

Can I reappear somewhere else?

Yes, describe the new place in the prompt of the third version. The more the new place resembles the old one, the cleaner the reappearance.

Is the sound included?

Yes, the node outputs the video with its audio. On a disappearance the small displacement of air does a good part of the effect, and you can replace it in the edit.

How long should my clip be?

Between one and fifteen seconds. Five are enough, and a shorter clip leaves the model fewer chances to make you move before the disappearance.

Does it work with several people?

Yes, and making one person out of two disappear gives a stronger result than a disappearance on its own. Name clearly who goes and who stays in the prompt.

Disappear in a video with AI

Use this workflow

Free account, no card required. The workflow opens pre-filled in your canvas.