01 · Quick answer
Start with the video you want to receive
New trio performance
Choose a dedicated trio re-shoot such as this site, Motion Transfer, TryH3, or GangnamStyleAI.org. Compare their photo and empty-slot rules before uploading.
One-person horse dance
Choose a dance-from-photo app whose current preview shows the 2012 horse-riding move. Do not assume every Gangnam Style page makes the same trend.
Existing footage or editing
Use a face swap when fixed bodies and footage are acceptable. Use CapCut to arrange your clips or add captions and publishing edits to a generated result.
Need a deeper template decision? Read the CapCut template vs AI comparison.
02 · Workflows
Four workflows can sit behind the same search phrase
Dedicated trio re-shoot
This site, Motion Transfer, TryH3, and GangnamStyleAI.org create new people and frames around a prepared trio sequence. Each service handles photo slots differently.
Dance from photo
A prepared motion animates one subject. Many pages using “Gangnam Style” in the title show the older horse dance rather than the three-person selfie sequence.
Face swap
Existing footage supplies the body, clothing, camera, and scene. This is fast, but the user does not control the complete generated appearance in the same way as a re-shoot.
General reference-video model
A general model gives more control over references and prompting, but also requires more setup. Model rules, audio, cost, and role consistency must be checked for each run.
03 · Comparison
Compare named options using current product-page facts
| Option | Workflow | Input | Output | Access and price | Best fit | Watch for |
|---|---|---|---|---|---|---|
| Gangnam Style AI Video (this site) | Dedicated trio re-shoot | One required Center portrait; optional Right and Left portraits. Empty side slots use preset neutral adults. | 15-second, 16:9, 720p MP4 with reference audio. | Sign-in required. The current Single pack is $9.99 and covers one 22-credit 720p generation. | A fixed Center → Right → Left → Center trio sequence with scene choices and no prompt writing. | Faces and details can vary. A completed retry uses credits again; third-party music rights are not included. |
| Motion Transfer | Dedicated three-photo trio re-shoot | Three separate portraits assigned to middle, screen-right, and screen-left. | A 15-second landscape video with sound; the page displayed a 480p setting during review. | Sign-in required. A live credit quote is shown before generation; 202 credits were displayed during review. | Users who already have all three portraits and want explicit role assignment. | The prepared street sequence and camera path are fixed; review the current quote before submitting. |
| TryH3 Gangnam Style AI Trio | Multi-model trio generator | One to three separate portraits. Its current page says Photo 1 repeats into empty positions. | 15-second, 16:9 MP4 with music; resolution choices depend on the selected model. | Account and credits required. The current page lists generation from 30 credits and refunds failed generations. | Users who want to choose among several models and resolution tiers. | Its empty-slot behavior differs from this site, and a higher resolution does not guarantee better likeness. |
| GangnamStyleAI.org | Dedicated trio re-shoot | Its current page describes one-photo or three-photo paths for the trio format. | The page lists a 15-second, 16:9, 720p video with a song under it. | The page listed $9.90 per video and no account when reviewed. | Users comparing a no-account single-purpose trio workflow. | Its role behavior, refund terms, and product claims belong to that service and do not describe this site. |
| CapCut media-replacement template | Prepared edit using existing photos or clips | Varies by listing. Two checked examples required either two clips at 9:16 or one clip at 3:4. | Uses the listing's prepared timing, crop, effects, and available audio rather than this site's fixed trio output. | Template availability and workflow can vary by platform, region, app version, and listing. | Arranging existing footage or editing an AI-generated MP4 after generation. | A traditional template does not automatically create the three-person performance from portraits. |
Prices, credit quotes, model choices, and account rules can change. The values above describe the pages observed on October 27, 2026; always check the live submission screen before paying.
04 · Evidence
A documented site result and a specification-based comparison
What this site observed
One three-photo run used the Beijing night setting and Seedance 2.0 Standard. It produced a 14.869-second 480p video and followed the fixed Center → Right → Left → Center role sequence.
How to read this comparison
This page compares published workflows, inputs, outputs and current examples. It does not declare a quality, speed, likeness or value winner from specifications alone.
The observed 480p feasibility sample is not the current standard product delivery. This site's current output is a 15-second, 16:9, 720p MP4. Resolution alone does not guarantee stronger face consistency.
05 · This site
This site uses a fixed human-trio sequence
- Center is required, appears in the first solo, and returns at the end.
- Right and Left are optional and control the second and third solo turns. Empty sides use preset neutral adults.
- Scene choices are Seoul, Convenience store, Office overtime, Beach wedding, Hotel lobby, or a custom text background.
- The result is a 15-second, 16:9, 720p MP4 with reference audio for rhythm and mouth movement.
- The tool does not clone a user's voice, grant third-party music rights, or guarantee exact face consistency.
Read how empty positions work in the one-photo guide, assign specific people with the friends guide, or diagnose identity drift in the face-matching guide.
06 · Cost and risk
Check the live price, retry rule, and usage rights
Account and current price
This site requires sign-in before submission. The current Single pack costs $9.99 and includes 22 credits for one 720p generation. One, two, or three portraits do not change the per-video price.
Check current pricingFailure and retry cost
Credits are reserved at submission and released if the task fails or the provider blocks it. A completed video that you choose to run again is a new generation and uses credits again.
Read the Refund PolicyPhoto and music rights
Use only your own portrait or an authorized adult's portrait. The reference audio helps the performance keep rhythm, but the export does not grant rights to third-party music.
07 · Method
How this comparison was checked
Reviewed October 27, 2026. This site's roles, scenes, output, audio, price, credits, failure behavior, and observed example were checked against its product specifications and documented example. Competitor inputs, outputs, account rules, and displayed prices were taken from the linked public product pages on the same date. No cross-app generation was purchased or run.
08 · Questions
Gangnam Style AI apps FAQ
Which app makes the 2026 three-person selfie trend?
Choose a dedicated trio re-shoot or a reference-video workflow that assigns portraits to the group and solo sequence. This site, Motion Transfer, TryH3, and GangnamStyleAI.org publish trio-specific workflows, but their input rules and empty-slot behavior differ.
What happens if I upload one or two photos on this site?
One Center photo is required. If Right or Left is empty, that side uses a preset neutral adult. With two photos, Center and the side you choose use your portraits while the remaining side uses a preset adult. The same face is not copied into every position.
Which app should I use for the 2012 horse dance?
Use a dance-from-photo app or a horse-dance template whose preview clearly shows the 2012 move. This site makes the newer three-person selfie sequence, not the horse-riding dance from the music video.
Does this site require an account?
Yes. You can inspect the upload form before signing in, but you must sign in before submitting a generation. Other products have different account rules, so check the current product screen.
What does one video cost on this site?
The current Single pack is $9.99 and includes 22 credits, enough for one 720p generation. Pricing is per generated video rather than per uploaded person, and a completed retry uses credits again.
Does the MP4 include sound?
Yes. This site's final MP4 includes the reference audio used for rhythm and mouth movement. It does not clone the user's voice or grant rights to third-party music.
What happens if generation fails?
This site reserves credits when a task is submitted. If the task fails or the provider blocks it, the reserved credits are released. A completed result that you choose to redo is a new generation and uses credits again.
Has this site tested a real three-photo output?
Yes. One observed run used three photos, the Beijing night setting, and Seedance 2.0 Standard, producing a 14.869-second 480p video. That confirms one real run and its fixed role sequence, not a cross-app quality winner or a guaranteed future result.
Can I edit the generated video in CapCut?
Yes. Download the MP4, then use CapCut or another editor for captions, crops, pacing, or publishing adjustments. Editing the file afterward does not change which generator originally created the people and performance.
Which app has the best face consistency?
Face consistency depends on the source photos, role mapping and generation model. Compare each app's current examples, resolution, retry cost and refund policy instead of choosing from specifications alone.
09 · Next step
Use the fixed trio workflow
Upload one Center portrait, add optional Right and Left portraits, choose a scene, and review the 22-credit 720p submission before you generate.
Open the Gangnam Style AI Video tool