Turn two photos into a 14-second NEON-inspired duo dance video. One photo takes the opening close-up and dances in front, the other steps in beside β vertical 9:16, reference audio kept.
Upload two photos, one per dancer. Clear, front-facing images with the face visible give the cleanest result. Dancer 01 opens the clip; Dancer 02 steps in beside.
By uploading, you confirm you're 18+, have permission to use this photo (no celebrities), and agree to our Terms of Service and Privacy Policy.
A2O Lite Neon AI refers to AI-generated videos in which two people, taken from photographs, perform the opening duo dance of NEON, the debut single of the K-pop girl group A2O LITE, released on 13 October 2026. The name is also written A20 LITE with a zero; both spellings mean the same group and the same dance.
The dance is a two-person intro that opens the group's debut stage. One dancer begins alone in a front-facing close-up, the second steps in beside her, and the two then move through the routine in tight sync before the camera closes in again. The whole intro runs about fourteen seconds and is choreographed by J Black with Cream Dance. The move most often clipped is a turn on one foot inside the synced section.
NEON was released on 13 October 2026 and reached No. 1 on Melon's TOP100 about 29 hours later. The intro travelled past the group's own audience for two reasons: the English hook "Neon and Beyond" is heard, in Korean, as a phrase about staying unmarried, and the mishearing became a joke in its own right; and the routine is short, tightly framed and easy to recognise, which made it simple to copy.
AI versions followed the jokes. Early examples replaced the two dancers with film and cartoon characters; later ones used photographs of ordinary people. What the AI versions change is the requirement: the two performers no longer need to share a room, or be dancers at all. Any two photographs will do.
This page is one of those generators. It takes two photos β one for the dancer who opens the clip and stays in front, one for the dancer who steps in beside β and returns a 14-second vertical video built from the same routine, with the reference audio kept. It is an independent tool and is not affiliated with, endorsed by or sponsored by A2O Entertainment or A2O LITE.
Filming the NEON duo intro yourself means two people, one room, one free evening and enough patience to match a short routine on camera. Most people who want the clip never get that far.
Two photos collapse that into a single upload. The routine is already fixed, so the only decision left is who stands in it β and because the two slots are separate, the pair does not have to exist in real life at all.
Choose a clear photo of the person you want in the opening close-up. They also dance in front for the rest of the clip, so pick the face you want seen first.
Add a separate photo for the dancer who steps in beside. Front-facing and evenly lit works best; if one photo is a selfie and the other a wide shot, the two bodies can come out at different scales.
Choose 480P, 720P or 1080P, check the credit cost, then generate. Rendering continues in the background if you leave the page, and the clip lands in My Videos as a 9:16 MP4.
Upload two separate photos, with one intended dancer in each image.
A 14-second vertical clip built from a fixed reference routine, with the reference audio retained. Three quality tiers, one MP4 out.
The full opening duo intro, first frame to last β not a trimmed highlight.
Fills the TikTok, Reels and Shorts frame. No letterboxing and no re-crop before you post.
Pick the tier before you generate. The routine, framing and length are identical at every tier.
The track from the reference clip is kept in the render. We add no music or voiceover, and grant no rights to that audio.
One file, ready to upload as it is.
Each slot maps to one character in the reference routine β Dancer 01 in front, Dancer 02 beside.
The movement comes from the reference clip, so there is nothing to describe in text and nothing left to guesswork.
480P for 45 credits, 720P for 90, 1080P for 180. Choose before you generate.
Start a render and close the tab if you want. The finished clip appears under My Videos.
One-time credit packs, no subscription. A video costs 45 credits at 480P, 90 at 720P and 180 at 1080P. Purchased credits never expire.
50 credits
Google sign-in required at checkout
140 credits
Google sign-in required at checkout
280 credits
Google sign-in required at checkout
A2O Lite Neon AI is a two-photo video generator built on the opening duo routine from NEON, the debut single of K-pop group A2O LITE. You supply one photo per dancer and get a 14-second vertical clip in which both people perform that routine.
The official spelling is A2O LITE, with the letter O. A20 LITE with a zero is a common misspelling of the same group and the same song, and both spellings point at the same dance.
It is the two-person opening of the NEON choreography: one dancer starts alone in a front-facing close-up, the second steps in beside them, and they finish the routine in sync. It is short, tightly framed and instantly recognisable, which is why it became the section people remake.
Two, one per dancer. Dancer 01 opens the clip in the close-up and dances in front; Dancer 02 steps in beside. Put the person you want seen first in the first slot β that is the only placement decision you make.
Front-facing, evenly lit photos with the face clearly visible. Half-body or full-body shots match the reference framing better than a tight face-only crop, and sunglasses, masks or heavy filters will cost you likeness.
No. The movement comes from the reference routine rather than from you, so nothing is filmed and nothing is performed. Your only input is two still photos.
Three tiers: 480P for 45 credits, 720P for 90 credits and 1080P for 180 credits. The routine, framing and length are the same at every tier; only the resolution changes.
14 seconds, exported as a 9:16 vertical MP4 with no watermark. That is the native shape for TikTok, Reels and Shorts, so it fills the frame without bars or a re-crop.
The audio track from the reference clip is retained in the render, and we add no music or voiceover of our own. That does not grant rights to the track itself, so check the platform's music rules before you post.
The slots take whatever you upload, and pets and illustrated characters are common. Results depend on how clearly the subject reads as a figure, so a clean, well-lit pet photo works better than a busy one.
No. This is an independent tool. It is not affiliated with, endorsed by or sponsored by A2O Entertainment, A2O LITE or their members, and it does not use the artists' faces, voices or likeness.
The render is yours to post. The routine, the audio and any recognisable person in it carry their own rights, so treat it as fan-made content and credit the trend rather than presenting it as official material.
Rendering continues in the background if you leave, and the clip appears under My Videos. A failed generation does not consume the credits for that video, so you can adjust the photos and try again.
Two photos, one fixed routine, fourteen seconds out. Pick the quality tier, generate, and download the clip.