The generator is for couples and friends who want this trend without filming it together, or who cannot hold the pose: Photo 1 takes the person with the phone, Photo 2 the person on the shoulders, and the prompt asks the model to follow a real couple's 10-second template clip with your faces. A video costs 78–431 credits (about $2.90–$16.02 at the lowest credit price); the 3 sign-up credits do not cover one. Output is 9:16, 480p to 1080p, watermarked until your first purchase.
How to make the me and my girlfriend AI trend
- 01
Add one photo of each person
Photo 1 is the person who holds the phone at the mirror; Photo 2 is the person who lies across their shoulders and pops out at the end. Any genders, swap them with one tap. One adult per photo, face sharp and turned to the camera.
- 02
Pick the template clip, the place and the resolution
Choose the elevator mirror or the Christmas living room, keep that place or describe your own, then pick 480p, 720p or 1080p. The button shows the credits before you start.
- 03
Check, download and post
Watch the moment the second face appears, download the 10-second vertical MP4 with the sound already on it, and turn on the platform’s AI label when you post.
Template clips your photos replace, and the move itself
These are template clips, filmed by real couples. The first two are the clips the generator follows: pick one in the tool and the prompt asks the model to recreate that pose, camera and sound with your two faces. The third is another couple doing the same move in a different room, so you can see how the trend varies.
The elevator template clip, 10 s: a man in a grey patterned tracksuit films the elevator mirror while his partner lies across his shoulders, her arms and legs hanging like extra limbs, then she lifts her head beside his face. Real people, with the clip’s own sound.
The Christmas template clip, 10 s: an older man in Christmas pyjamas films a tall living-room mirror while his partner lies across his shoulders in matching pyjamas, then she pops out beside his face, under a garland and stockings. Real people, with the clip’s own sound.
Another couple’s take on the same move, 11 s: she hangs over his shoulders from behind in a bright hallway while he films a round mirror, then she leans out beside his face. A different room and hold from the two template clips, shown so you can see how the trend varies. Real people, with their own sound.
What is the me and my girlfriend AI trend?
The me and my girlfriend AI trend is the AI version of a TikTok and Reels mirror selfie: one person films themselves in a mirror while their partner lies sideways across their shoulders, so the hanging arms and legs make it look like one person with four arms, and then the partner's face appears beside them. The AI version builds that clip from two separate photos instead of filming it. It has nothing to do with AI girlfriend chat apps.
If you only remember the picture, it is the mirror selfie where someone seems to have four arms because their partner is lying across their shoulders; the same trend also goes by me and my girlfriend mirror selfie AI, or simply the me and my girlfriend trend. The caption "Me and my girlfriend" comes from an elevator mirror version posted to TikTok on 6 November 2026; remakes in living rooms, hallways and Christmas pyjamas followed. Filming it for real takes a strong pair of shoulders and a few tries; the AI route needs no lifting and no shared location, but a generated video can show wrong hands, a merged arm or a face that changes at the reveal, which is how viewers tend to spot AI versions.
The me and my girlfriend shot, second by second
Both template clips run 10 seconds and follow the same three beats; the table shows what each photo does in each beat, so you can choose photos that suit the role. If you only want a dance rather than this pose, the AI couple dance video generator takes the same two photos.
| Time | What happens | What viewers see |
|---|---|---|
| 0–4 s | Photo 1 stands at the mirror with the phone at chest height. Photo 2 is already lying sideways across their shoulders: legs hanging on one side, head and arms on the other. | At a glance, one person with extra arms and legs. |
| 4–7 s | In the Christmas clip Photo 2’s face comes into view beside Photo 1’s shoulder; in the elevator clip it is turned down at the side from the start. | Viewers work out that there are two people. |
| 7–10 s | Photo 2 lifts their head and smiles into the mirror next to Photo 1, who keeps filming. | The reveal: both faces in the reflection. |
| Template clip | Place and camera | When the second face shows | Good for |
|---|---|---|---|
| Elevator mirror | Brushed-steel elevator, full-height mirror, phone at chest height | Turned down at the side from the start, lifts at the end | The look most posts copy; everyday clothes |
| Christmas room | Living room with a garland and stockings, tall mirror | Hidden behind Photo 1's head, appears around 4 s | Holiday posts, a clearer surprise |
| Custom | Your words, plus an optional place photo; still a mirror selfie | Follows the elevator or Christmas clip you picked | Bathrooms, hotel rooms, your own home |
Is the me and my girlfriend AI trend free to make here?
No, the me and my girlfriend AI trend is not free to make on Genjutsu AI: one video costs at least 78 credits (about $2.90), and the 3 sign-up credits are not enough on their own. Watching the template clips and checking every price is free, and the button shows the exact credits before anything is charged. From a zero balance, the smallest purchase that covers one video is $12 for 220 credits, with no subscription. Videos from accounts without a purchase carry a moving watermark; any purchase unlocks clean downloads.
| Question | What happens here |
|---|---|
| Price per video | 78 to 431 credits, set by the resolution (about $2.90 to $16.02 at the lowest credit price). Both template clips and the custom place cost the same. |
| Free credits | 3 sign-up credits, valid for 7 days. Not enough for a video on their own: the cheapest is 78 credits. Any credits count towards it, gifts included. Watching the template clips is free. |
| Smallest purchase | The $5 Trial pack (24 credits) is not enough for one of these videos. The smallest pack that covers one is $12 (220 credits). One-time packs, no subscription needed. |
| Watermark | Videos from accounts without a purchase carry a moving site watermark. Any purchase unlocks clean downloads of earlier and later videos. |
| Length and format | 10 seconds, the whole template clip. 9:16 vertical MP4; 480p, 720p or 1080p on Seedance 2.5. |
| Sound | The template clip’s own sound is laid back on the finished video. There is no song picker; swap the audio in your editor or pick a sound when you post if you want another song. |
| Failed or rejected | If the model fails or times out, returns a broken file, or a photo is turned down by the safety check or the model’s review, the credits go back automatically. If the model delivers and only the final export on this site fails, the charge stays and support finishes the video from the saved output. |
| Finished but not right | A completed video is not refunded. Another take is charged again at the price on the button. |
Mirror selfie video price and credits
One 10-second video is 78 credits at 480p, 175 at 720p and 431 at 1080p on Seedance 2.5, the only model offered here because the template clips are real people. Both template clips and the custom place cost the same; Seedance 2.5 bills per second for the clip it follows and for the finished video.
| Model · length | 480p | 720p | 1080p |
|---|---|---|---|
| Seedance 2.5 · 10 s | 78 (≈$2.90) | 175 (≈$6.51) | 431 (≈$16.02) |
Credits come from the same function the generator charges; the dollar figures use the lowest credit price (a monthly plan), so one-time packs work out a little higher per credit. Packs and plans are on the pricing page.
What the mirror selfie generator gets wrong
This generator has not been tested on paid runs yet, so read these as the usual weak spots of two-person, reference-guided video: the prompt asks the model to keep each face from its own photo, but results vary with the model and the photos. It makes two people only, keeps the 10-second length, and cannot swap the pose for another trend.
| What you might see | Why | What helps |
|---|---|---|
| Photo 2's face looks different when it appears | The face is hidden or turned for most of the clip, so the model has little to hold on to | A sharp, front-facing Photo 2 in even light; the elevator clip, where the face shows earlier; another take |
| Extra or merged hands, or the phone disappears | Four arms around one body is exactly what video models get wrong | Photos with arms free and hands visible; 720p or 1080p |
| Something of the template couple shows through | The template is a real filmed clip | Clear, well-lit photos that fill the frame |
| Very different heights look odd | The pose comes from the template couple’s proportions | Photos taken from a similar distance and angle |
| A custom place looks off | The model redraws the whole place around the same mirror shot | A short, concrete description and a place photo |
| A pet instead of a partner | Not offered: the pose needs a person lying across the shoulders | Use two people |
| Seedance 2.5 turns a photo down before it starts | The provider’s review refused the photo; the credits come back | A clearer photo of one adult |
Use only your own photo and photos of adults who agreed to be in the video, and no children. For other two-person videos, this site also has the AI couple dance video generator and the AI love story video.
最近更新: 2026-11-20