すべてのトレンドを見る

Mini Me AI Video

Upload one photo and meet the little you everyone is posting about: a child version of you, in a mini copy of your outfit, walks in, stands beside you and strikes your pose. 8 seconds, $6.99, no account, and a free 3-second preview before you pay.

別名:mini me filtermini me trendmini version of me AI

  • $6.99・一度きりのお支払い
  • アカウント不要
  • 音声付き8秒の動画
  • 失敗時は全額返金
  • 透かしなし

動画を作る

写真1枚をアップロードして、お支払いは一度だけ。物語の部分はこちらで用意します。

アカウント不要。生成に失敗した場合は全額返金します。

使用例

またはプロンプトスタジオで自分だけのバージョンを

こちらは別のカスタム制作コースです。好きなシーンを書き、好きな長さを選び、クレジットで支払います。上の $6.99 の動画は完成品プランで、あなたの写真と固定の物語、アカウント不要で作れます。

そのまま使えるMini Me AI Videoのプロンプト集

Side by side, same outfit
“Use the uploaded photo as the opening frame: the adult in it stands facing the camera. A small child version of the same person, about six years old, with the same face shape, eyes, smile, hair color and skin tone, walks in wearing a child-sized copy of the adult’s outfit and stops beside them. 0-3 s: the child walks in. 3-5 s: the adult looks down and laughs. 5-8 s: both strike the same pose at the camera and the child waves. Preserve the original face. Static camera, 9:16, no text.”
If I had a mini me
“An adult sits at a café table and lifts a cheerful toddler, about two years old, onto their lap. The toddler has the same eyes, smile, hair color and skin tone as the adult, like a tiny copy, and wears a little version of the adult’s sweater. The toddler reaches for the adult’s face, both laugh, and they look at the camera together. Warm window light, handheld phone feel, 9:16, wholesome, no text, no logos.”
Mini me on the porch
“An older woman sits in a garden chair on a sunny porch with flowers. A little girl of about five with the same face, the same smile and the same cardigan in a child size walks up holding an orange and sits in the small chair beside her. They look at each other, laugh, then both smile at the camera. Soft afternoon light, static shot, photorealistic, 9:16, no text.”

Mini Me AI Videoとは?

Mini Me AI is the name people search for a short video in which they appear next to a small child version of themselves. The child has their face, their smile and their hair, and wears a tiny copy of their outfit, so the two look like the before and after of the same person standing in one frame. It started as a CapCut template and an AI filter on TikTok, with captions like “this effect reveals your mini version” or “meet your mini me”, and it has been one of the most used AI templates of the year.

It works because everyone is curious about it. People wonder what they looked like as children, parents wonder what their child would have looked like if they took after them completely, and friends love comparing the result with real childhood photos in the comments. The matching outfit is the visual joke: a toddler in a tiny leather jacket or a little business blazer looks sweet and a little absurd. Grandparents with a mini me in a matching cardigan are some of the most shared versions.

On this page there are two ways to make one. The ready-made video at the top takes one photo of you and gives you the tested 8-second moment: your mini me walks in, you laugh, and you both strike the same pose, with no account. Further down, the prompt studio lets you write your own version: a baby in your arms, a different place, a longer scene or a growing-up timeline. Either way, use a clear photo, keep the outfit visible, and post it in 9:16 with the trend sound added in TikTok or Reels.

プロンプトをコピー

“A warm, cheerful clip filmed like a phone video, photorealistic. An adult stands facing the camera in a bright everyday setting and smiles. A small child version of the same person, about six years old, with the same face shape, eyes, smile, hair color and skin tone, wearing a child-sized copy of the adult’s outfit, walks in and stops beside them. The adult laughs in surprise, then both turn to the camera and strike the same pose, and the child takes the adult’s hand and waves. Both faces visible, 9:16, no text, no logos.”

01

What you upload and what you get

One photo in, one finished Mini Me AI video out. There is nothing to write, nothing to tune and no account to create.

You upload one clear photo of yourself, waist-up or full-body, as JPG, PNG or WebP, up to 10 MB each. We turn it into a single 8-second video in 720p with natural scene sound, in 9:16 for TikTok, Reels and Shorts or 16:9 for a wider picture. There is no watermark, and the file is yours to post.

You bring
One photo with a clearly visible face, and a tick to confirm the person in it is an adult who agreed to be in the video.
You get
One MP4, 8 seconds, 720p, with sound and no watermark, on a private order page you can open for 14 days.
It takes
Usually 5 to 10 minutes. You can leave the page and come back through the same link; the video keeps rendering without you.
It costs
$6.99 once, paid before the render starts. No subscription and no extra charge for the download.

02

The story your photo becomes

The story is fixed on purpose. Every order gets the same tested moment, so the result looks like the Mini Me trend you came for, not a random experiment.

The video opens on you, in the outfit and the kind of place from your photo, smiling at the camera. Then a small child walks in from the side and stops right next to you. It is you at about six years old: the same face shape, eyes, eyebrows, smile, hair color and skin tone, and a child-sized copy of exactly what you are wearing. You look down, laugh in surprise, and then the two of you turn to the camera and strike the same pose side by side. In the last second your mini me takes your hand and gives a small wave, with both faces clear and in focus.

Your mini me is not taken from your childhood photos. The AI works out what you probably looked like as a child from the face in the photo you upload, the same way the CapCut Mini Me filter does, so think of it as a very good guess rather than a record. People who post it usually say the eyes and the smile are what make it land. The outfit matters more than you might expect: whatever you wear in the photo, your mini me wears too, so a bold jacket, a team jersey or a bright dress makes the matching look obvious at a glance.

写真: Waterfront at dusk: the cream jacket and floral dress come back in child size, and the hand-hold and wave land on time.写真
動画
Waterfront at dusk: the cream jacket and floral dress come back in child size, and the hand-hold and wave land on time.
写真: Café: the denim jacket and white T-shirt are copied, with the same curly hair and the same grin.写真
動画
Café: the denim jacket and white T-shirt are copied, with the same curly hair and the same grin.
写真: Porch: the mustard cardigan in child size, and the grandmother’s smile carried over to her mini me.写真
動画
Porch: the mustard cardigan in child size, and the grandmother’s smile carried over to her mini me.

Real results from test orders, each sent exactly as a buyer’s order is: one photo in, the 8-second video out, no editing. The people in these photos are AI-generated, not customers.

03

Which photos work best for a Mini Me AI video

The video is only as good as the face it starts from. A minute spent choosing the photo saves a lot of disappointment.

Face clearly visible
Your mini me is built from your face, so pick a photo where it is sharp, well lit and looking at the camera or slightly turned. The clearer the eyes and the smile, the closer the resemblance.
Waist-up or full-body
A photo that shows your outfit works best, because the child wears a mini copy of it. A tight face-only selfie leaves the AI guessing what you are wearing.
Just you in the photo
Use a photo of one adult. If there are several people, we cannot tell whose mini me to make, and other faces can leak into the child.
No sunglasses or heavy filters
Sunglasses, masks, beauty filters and big stickers hide the features your mini me inherits, so the child looks less like you. Natural light and your everyday look work best.

04

Choose 9:16 for Reels or 16:9 for a wider story

Pick the shape before you pay. It is part of the order and cannot be changed after the render starts.

Choose 9:16 if you will post to TikTok, Instagram Reels or YouTube Shorts: the clip fills the phone screen and the face stays large. Choose 16:9 if you want a wider scene, for example for YouTube, a slideshow or a TV, where the surroundings have more room.

05

How delivery, downloads and privacy work

There is no account, so your private order link is the key to your video. Keep it.

Right after payment you land on your order page. It shows the progress, then plays the finished video and gives you a download button. The page stays available for 14 days, and this browser also remembers your recent orders. Anyone who has the link can open the video, so do not share it publicly.

Your photo is used only to make this one video. It is deleted as soon as the video is finished, or when the order is refunded. The finished video is deleted when the 14 days are over, and the order page then tells you it has expired. Order pages are not indexed by search engines.

06

Mini Me AI video pricing

One price, one video. You pay before the render starts, and you are refunded automatically if we cannot deliver it.

WhatDetails
Price$6.99 one-time, in USD, paid by card through Stripe
Video8 seconds, 720p, with natural sound, no watermark
Formats9:16 vertical or 16:9 wide
AccountNot needed
DeliveryPrivate order page, available for 14 days
If it failsOne automatic retry, then a full refund to your card

07

Baby, toddler and growing-up versions

The paid video above shows a school-age mini me standing beside you. Holding a baby version of yourself, or watching yourself grow up, is a different shot that you can make with your own prompt.

The Mini Me trend has three popular looks. The side-by-side version, with a matching outfit and the same pose, is what the ready-made video makes. The “if I had a mini me” version shows you holding a baby or toddler with your face, and the growing-up version turns one face from child to adult in a single shot. For the last two, use the prompt studio further down this page: the prompt cards include a baby-in-arms version, and the age progression page has prompts written for a full growing-up timeline.

08

Key features of the prompt studio for a custom mini me

For a version of your own, with a baby, a different place, another age or a longer scene, paid with credits. These are the parts the studio handles for you.

Your photo as the opening frame
Upload a clear photo of yourself and it becomes the first frame, so the adult in the video starts from a real picture and the child is built from the same face. Add “preserve the original face” to keep you looking like you.
A ready-made prompt in the composer
The prompt at the top of this page is already loaded in the composer. Use it as it is, or change the age, the place and the pose. Every prompt on this page drops in with one click.
Any age for the mini me
Write the age you want, from a baby in your arms to a ten-year-old. Name it as a number: “about two years old” gives a far more consistent child than “young” or “little”.
Longer scenes and more beats
Seedance 2.5 renders up to 30 seconds, so your mini me can walk in, play, copy a few of your moves and wave goodbye. Give each beat two or three seconds and keep both faces in frame.

09

The three beats of an 8-second mini me reveal

This is the moment the ready-made video uses, and it is a good base for your own. Name the time of each beat in the prompt.

TimeWhat happensHow to write it
0–3 sThe mini me walks in“a small child version of the same person walks in from the side and stops beside them”
3–5 sThe surprise“the adult looks down and laughs in surprise”
5–8 sSame pose, a wave“both turn to the camera and strike the same pose; the child takes the adult’s hand and waves”
End on both faces at the camera. A clip that stops while the child is still walking feels cut off.

10

How to make the mini me look like you

The resemblance is what people comment on, so it is worth getting right. A video model cannot see your childhood; it can only follow your face and your words.

Name the features the child should share, one by one: face shape, eyes, eyebrows, smile, hair color and skin tone. A prompt that only says “a child who looks like them” leaves the model free to invent a generic child. Saying “as if taken from their own childhood photos” helps it keep the proportions of a real child rather than a shrunken adult.

Name the features
Eyes, smile and hair carry most of the likeness. List them in the prompt so the model keeps them.
A child, not a small adult
Give an age and ask for a child’s proportions: a rounder face, bigger eyes, a shorter body. Without it, some renders look like a miniature grown-up.
Same outfit, child size
Say “a child-sized copy of the adult’s outfit”. The matching clothes tell viewers who the child is before they study the face.
Faces to camera at the end
Finish with both faces turned to the camera and in focus. That last frame is what people pause on and screenshot to compare.

11

Outfits and places that suit the trend

The look sells the joke before anyone moves. A recognizable outfit and a simple place keep the eye on the two of you.

Bold, simple outfits
A red dress, a denim jacket, a suit or a team jersey reads instantly in mini size. Busy patterns blur on a small child.
City lights at dusk
A skyline behind you gives the glamorous “mini me effect” look from the original templates.
Home, café or porch
An everyday place feels real and warm, and suits the “if I had a mini me” version best.
Studio backdrop
A plain grey or white backdrop turns it into a portrait shoot and puts all the attention on the matching look.

12

Settings by where you will post

Decide the destination before you render. It changes the ratio and how much of the two of you shows.

WhereAspect ratioWhat works
TikTok, Reels, Shorts9:16You in frame from the first second, the mini me walking in by second three, the trend sound added in the app
Family group chat9:16Send it next to a real childhood photo and let everyone compare
Instagram feed4:5A closer frame so both faces and the matching outfit fill the picture
YouTube or a slideshow16:9A wider scene with room for the child to walk in from the side

Mini Me AI Videoの作り方

  1. 1

    Add your photo or pick a prompt

    Upload one clear, well-lit photo of yourself, waist-up or full-body, with your face and outfit visible. Only use a photo of an adult who agreed to it. Or skip the photo and start from one of the prompt cards above.

  2. 2

    Set the age, the outfit and the beats

    The trend prompt is already in the composer. Write the mini me’s age as a number, name the features it shares with you, ask for a child-sized copy of your outfit, and end with both faces at the camera.

  3. 3

    Generate, check the likeness and download

    A clip usually takes a few minutes. Check that the child looks like you and that both faces end in focus, retake with one change if not, then download in 9:16 and add the trend sound when you post.

Mini Me AI Videoのアイデア

Grandparent and mini me

An older face next to its five-year-old self is the version people share most.

Office twins

A tiny you in a suit at the next desk is the funniest workplace version.

Matching gym session

Your mini me copies your warm-up in the same tracksuit.

Kitchen helper

Cooking together in matching aprons feels warm and unplanned.

City night glam

The skyline version from the original templates.

Mirror reveal

You look in the mirror and the reflection is your mini me.

Mini Me AI Videoでよくある失敗と、その直し方

失敗例

The child looks like a stranger

直し方

Use a sharp, front-facing photo and list the shared features in the prompt: face shape, eyes, eyebrows, smile, hair color and skin tone. Add “as if taken from their own childhood photos”.

失敗例

The mini me looks like a tiny adult

直し方

Give an age as a number and ask for a child’s proportions: a rounder face, bigger eyes, shorter limbs. “About six years old” works better than “small” or “mini”.

失敗例

The outfits do not match

直し方

Describe the outfit completely, top and bottom, and say “a child-sized copy of the adult’s outfit”. Use a photo where your clothes are visible, not a face-only selfie.

失敗例

Your own face changes

直し方

Add “the adult keeps the exact face from the reference photo” and keep the adult facing the camera most of the time. Avoid big turns and hands over the face.

その他のバズトレンド

プロンプトガイドハロウィンおばけ屋敷、カボチャパーティ、仮装シーンの見せ場。完成テンプレート · 写真をアップロードゾンビカップル恐怖、気づき、抱擁、そして幸せな思い出を8秒で。完成テンプレート · 写真をアップロード空飛ぶほうきほうきに乗って、お使いにも通勤にも、月夜の飛行にも。完成テンプレート · 写真をアップロードフルーツアーマーひとつの果物が、身にまとう鎧に変わります。完成テンプレート · 写真をアップロード動物タッチやさしく触れるだけで、動物の配色があなたの鎧になります。完成テンプレート · 写真をアップロード飛行車推進器を点火して、渋滞の車列をひと跳び。新着完成テンプレート · 写真をアップロードChristmas DanceTwo photos, matching pajamas, one 8-second duo dance by the tree.新着完成テンプレート · 写真をアップロードWassup in a MinuteThree photos, one misty hillside, one 8-second trio dance with your friends.プロンプトガイド巨大怪獣いたって普通の場所に、巨大な生物が現れます。プロンプトガイドアクションチェイス大作級のカメラワークで描く、スイング、ダイブ、フライバイ。プロンプトガイドミニチュアワールド水たまりやカップ、本棚の中にある小さな生きた世界。プロンプトガイドストップモーションポップアップブック、紙の生き物、手づくりの味わい。プロンプトガイドAI写真動画化静止した写真が瞬きし、動き出し、生き返ります。プロンプトガイドシュール動画完全に普通の場所で、ただ一つの不可能が起こります。プロンプトガイドToy to RealA toy on the table, then the real thing in the driveway.プロンプトガイドKnight Sword SwapA tiny knight lowers his sword. A giant guardian offers flowers.プロンプトガイドDolly ZoomThe subject stays put. The whole world behind it stretches.プロンプトガイドEarth ZoomFrom the whole planet to one front door, in a single dive.プロンプトガイドBirthday VideosCandles, confetti and one warm clip to send.プロンプトガイドProduct VideoOne product, one clean shot, ready for the listing.新着プロンプトガイドChristmasSanta visits, sleigh rides, snowy streets and festive reveals.新着プロンプトガイドFancy CafeSitting on the table, coffee in hand, until someone shows you how.新着プロンプトガイドAge ProgressionWatch one face grow from child to grandparent in seconds.新着プロンプトガイドThanksgivingDinner tables, autumn light and a thank-you worth sending.新着プロンプトガイドBaby DanceOne baby photo, one tiny dance, endless replays.新着プロンプトガイドPet DanceYour dog or cat, finally showing off the moves.新着プロンプトガイドRobot ArmMid-selfie, a robot arm reaches in and takes the phone.新着プロンプトガイドBicep HeartFlex, then turn both arms into one big heart.

その他のAI動画トレンドツール

以下の各トレンドページでは、写真やプロンプトから完成したショート動画を作れます。すぐ使えるプロンプト、プラットフォームごとの設定、よくある失敗の解説も付いています。

Mini Me AI Videoのよくある質問

2026年11月17日 に更新

  • It costs $6.99 once for one finished 8-second 720p video with sound and no watermark. You pay by card through Stripe before the render starts. There is no subscription and no extra fee for the download.

  • No. You upload one photo, tick the confirmation, pay, and land on a private order page. Your order link is the key to the video, so keep it. This browser also remembers your recent orders.

  • Usually 5 to 10 minutes after payment. The render runs in the background, so you can close the page and open your order link again later.

  • If the render fails, we retry it once automatically. If it still fails, the order is refunded in full to your card and the order page says so. If a render finishes but you are not happy with it, email support@dimly.art with your order link and we will look at it.

  • One clear photo of yourself, waist-up or full-body, well lit, with the face large and visible. Avoid sunglasses, masks, heavy filters and photos with several people. JPG, PNG and WebP up to 10 MB each are accepted.

  • They are used only to make your video and are deleted as soon as it is finished or the order is refunded. The finished video is deleted after 14 days.

  • Only if every person in the photos is an adult and has agreed to appear in the video. You confirm this with a tick before you pay. Photos of children, or of people who did not agree, are not allowed.

  • Yes. Choose 9:16 for TikTok, Reels and Shorts, or 16:9 for a wider picture. You pick it before you pay, and it cannot be changed once the render has started.

  • The ready-made video is 8 seconds long. For other lengths or your own scene, use the prompt studio further down this page, which is a separate path paid with credits.

  • The result is made to be non-graphic and suits TikTok, Instagram and YouTube. Platforms may ask you to label AI-generated content, so add the label they require when you post.

  • About six years old: small enough to look clearly like a child next to you, and old enough to walk in, copy your pose and wave. That age reads best on a phone screen. For a baby or toddler version, use the prompt studio below.

  • It will look like a child version of the face in your photo: the same eye shape, eyebrows, smile, hair color and skin tone. It is the AI’s best guess, not a copy of your childhood photos, so it may not match your real kid pictures exactly. A sharp, front-facing photo gets the closest result.

  • The video is rendered by ByteDance’s Seedance 2.5 video model, which dimly.art reaches through the Kie API. Your photo is sent there only to make your video, and our copy is deleted as soon as the video is finished or the order is refunded. The finished video sits on your private order page, which search engines do not index.

  • A render that fails is retried once and then refunded automatically. If the video is delivered but does not follow the story on this page, or you are not recognizable in it, email support@dimly.art with your order link within 14 days: we re-render it for free when that can fix it, and refund in full when it cannot. Your mini me’s face is the AI’s guess at your childhood look, so how closely it matches your real kid photos is not a defect; that is what the free preview is for. The full rules are on the refund policy page.

  • It makes the same kind of video: you and a little version of you in matching clothes, side by side. It does not use any CapCut template, creator video or song, and dimly.art is not affiliated with CapCut or TikTok. You do not need the app, and the faces in the video come from your photo.

  • It is a short video or photo in which you appear next to a small child version of yourself, with your face, your smile and a tiny copy of your outfit. It started as a CapCut template and an AI filter on TikTok, and people post it to see what they might have looked like as a child.

  • Yes. Start from the “If I had a mini me” prompt card, set the age to about one or two years old, and describe you holding the child on your lap. Keep it to a short, calm moment: lifting, laughing and looking at the camera render well.

  • Only with a photo of an adult who agreed to it. Ask them first, and use a clear photo of them alone. Never use photos of children or of people who did not agree.

  • A 5-second clip usually takes one to three minutes, and you can keep chatting while it renders. Seedance 2.5 goes up to 30 seconds if you want a longer scene, and MiniMax H3 renders 5 to 15 seconds.

  • You get free credits when you sign up, so you can try your first clips in the prompt studio without paying. After that, videos use credits, charged per second of finished video, and failed renders are refunded automatically.

  • No. dimly.art does not add a watermark, so you can post the download straight to TikTok, Reels or Shorts.

  • Add music in TikTok, Reels or your editor afterwards, and only use tracks you have the rights to; trend sounds are usually licensed for use inside the app only. All paid plans include a commercial license for the video itself.

  • いいえ。ブラウザでそのまま動きます——アプリのインストールもテンプレート探しも不要です。このページ自体がスタジオです:写真を追加し、プロンプトを貼り付けて、レンダリングするだけ。

  • ChatGPT はアイデアの静止画は作れますが、この動画は作れません。Mini Me AIの動画は画像から動画を作るモデルで生成され、このページが Seedance 2.5 を代わりに動かします。写真をアップロードして無料プレビューを確認し、完成した動画に一度だけ支払うだけ。プロンプトを書く必要はありません。

  • テンプレートは他人のクリップを再編集するもので、あなた自身をシーンに入れられません。このページはあなたの写真からあなたのバージョンをレンダリングします。有料クリップは透かしなしでダウンロードできます。

今日からMini Me AI Videoを作りましょう

まずは無料でスタート。最初の1本は、プロンプトひとつで作れます。