The Gangnam Style trend · 1 to 3 photos in
Gangnam Style AI
写真1〜3枚 · 7秒 · 9:16 · 音なし
あなたの写真
0/3
1枚だけならあなたが中央、両側はAIが作った友だちになります。友だちの写真を足すと本人が登場します。
無音動画 · 追加するサウンド: Gangnam Style – PSY
約3〜8分。失敗した場合はクレジットを返却します。
このエフェクトで作った画像です。あなたの動画には写真の人物が登場します。
Add your photo, and up to two friends. AI squeezes the three of you into one front-camera selfie on a Gangnam street at night, then films the rest: everyone sings along, you take the lens and point at it, the phone swings to a friend for a finger heart, and you all pile back in. 7 seconds, vertical and silent: add the trend's sound when you post.
- 1 photo works: AI adds two friends
- The trio selfie, passed between faces
- Silent, ready for the trend's sound
01Made with this generator
Gangnam Style AI Trio Selfies Made From Photos
A woman in a mustard cardigan, a man in a denim jacket and a woman in an olive puffer jacket
Made from the 3 photos below. The three sing to the lens, the middle friend points at it, the phone swings to the left for a finger heart. A silent 7-second sample.
A woman in a yellow cardigan
Made from the photo below, with two AI-made friends beside her. The three sing to the lens, the middle friend points at it, the phone swings to the left for a finger heart. A silent 7-second sample.
A man in a red track jacket and a woman in a camel coat
Made from the 2 photos below, with one AI-made friend on the right. The three sing to the lens, the middle friend points at it, the phone swings to the left for a finger heart. A silent 7-second sample.
02Before and after
From Your Photos to the Opening Selfie
Your photos
Opening frameA woman in a mustard cardigan, a man in a denim jacket and a woman in an olive puffer jacket
The three photos that went in (photo 1 in the middle, photo 2 on the left, photo 3 on the right), and the opening selfie drawn from them, every face and outfit kept.
Your photos
Opening frameA woman in a yellow cardigan
The photo that went in, and the opening frame drawn from it: the same face, hair and clothes in the middle of a three-friend selfie on a Gangnam street at night; the two friends beside are AI-made.
Your photos
Opening frameA man in a red track jacket and a woman in a camel coat
The photos that went in (photo 1 in the middle, photo 2 on the left), and the opening frame drawn from them; the friend on the right is AI-made.
03The trend
What Is the Gangnam Style AI Trend?
The name comes from PSY’s 2012 hit “Gangnam Style”. The 2026 trend is not the horse dance: it is the “gangnam style trio”, three friends sharing one phone’s front camera and taking turns singing the chorus into it. The reel that set off the 2026 wave, “gangnam style in gangnam” by @emiliyyam, was filmed at night in Seoul’s Gangnam district on 5 October 2026 and passed 17 million views; AI remakes with other faces followed within a week.
This generator makes that selfie from 1 to 3 photos. Photo 1 is always the friend in the middle, photo 2 the one on the left and photo 3 the one on the right; with fewer photos, the missing friends are drawn as AI-made strangers. An image model first draws the three of you in one front-camera selfie on a Gangnam street at night, keeping each face and outfit, then a video model films the clip. The clip is silent, so you add the trend's sound when you post.
The clip does not use the original recording, the music video or anyone’s reel, and the street is generated, not a real location shoot. Dimly is an independent tool and is not affiliated with PSY, his label or anyone connected to the song.
04How to make it
How to Make a Gangnam Style AI Video
- 01
Add 1 to 3 photos
Photo 1 is you, in the middle. Add a friend for the left and one for the right, or leave them empty and AI draws two friends. One person per photo, face to the camera; each is cropped to vertical 9:16.
- 02
Generate
The opening selfie is drawn first, then the clip is filmed from it. About 3 to 8 minutes; you can leave the page.
- 03
Download and post
A 7-second, 720p vertical MP4 without sound. Add the song in TikTok, Reels or CapCut when you post.
05Why this generator
The Gangnam Style AI Clip, Starring You
Your friends, or AI-made ones
Each photo becomes one friend in the selfie, with their own face, hair and clothes. Missing friends are drawn as strangers, so one photo is enough.
The real trio format
One arm-length front camera, three faces, the phone passed between them: the shape of the reel everyone copied, not a solo dance.
Silent, ready for the real sound
The clip has no audio track, so the trend's sound you pick in TikTok, Reels or CapCut plays clean over it.
Pay per video, refunded on failure
144 credits, shown on the button before you start. If a render fails, the credits go back automatically.
06Template or photo
Gangnam Style AI: This Generator vs. a Video Template
- Who is in the shotYou and up to two friends, from photosYou, filmed in an editing app
- What you need1 to 3 photosA video of yourselves and an app
- SoundNone: you add the trend's sound when you postDepends on the template
- Output7-second 9:16 MP4 at 720pDepends on the app
- Cost144 credits per videoPer the app's pricing
07Questions
Gangnam Style AI FAQ
Add one photo of you, and up to two friends if you like, then generate. AI draws the three of you in one front-camera selfie on a Gangnam street at night and films you taking turns singing to the lens: a 7-second silent vertical video, ready for the trend's sound.
No. The video is silent. The song belongs to its artists and label, so we neither include it nor imitate it. When you post, pick the trend's sound in TikTok, Reels or CapCut. Moves and mouth shapes are not timed to the original recording.
Download the MP4, upload it in TikTok, Reels or CapCut, tap Add sound and pick the trend's sound, then line its first beat up with the start of the clip. Posting with the trend's own sound also puts your video on that sound's page.
No. One photo is enough: you take the middle and AI draws two friends beside you. Add a second or third photo to put your real friends in, one person per photo. Photo 1 is always the middle.
No. The 2026 Gangnam Style trend is the trio selfie: friends taking turns singing into one front camera. The horse-riding dance from the 2012 music video is a different, older clip.
144 credits per video. The Weekly Pass ($6.99, 150 credits, cancel anytime) covers one, or a one-time $9.99 pack gives 250 credits. Failed renders are refunded.
Free to start: planning, a quote and your first 6 images. Video renders use credits, and this effect costs 144 credits. Failed renders are refunded.
A 7-second vertical (9:16) MP4 at 720p without sound, made for TikTok, Reels and Shorts, where you add the trend's sound.
No. Dimly is an independent tool. It is not affiliated with PSY, his label or the song, and it uses no original footage, recording, reel or logos.
You must upload only photos of yourself or of people who have agreed to appear. Do not upload photos of others without their permission.
08More effects
More Song Trend Videos
Other song-driven videos made from your photos.
