How to Make a When Two Legends Try to Make Coffee AI Video From 2 Photos
What’s new on this page
- — Published: our first When Two Legends Try to Make Coffee AI video (Wan 3.0, 15 s setting, 14 s clip, 480p, 300 credits, first try, about 6 minutes), the full prompt, screenshots of the template and the task, who goes where and photo tips.
A When Two Legends Try to Make Coffee AI video puts two people from two photos into the TikTok kitchen clip where a pair tries to make one coffee and gets every step wrong: grounds on the counter, a second scoop that misses, and milk poured where the mug used to be. We made the clip above on 2 November 2026 from two photos of our own. This is exactly what we did, what it cost, and how you can make the same thing with a friend.
What happens in a When Two Legends Try to Make Coffee AI video?
The camera stays at the end of a kitchen counter with one red mug, a kettle and a jar of coffee. Nobody panics; that's the joke. In our result:
| Seconds | What happens |
|---|---|
| 0–2 | Photo 1 stands in front with a spoon by the red mug; photo 2 stands behind and lifts the kettle |
| 2–5 | The first spoonful of coffee lands on the counter, not in the mug |
| 5–8 | The next scoop misses too, while the kettle keeps pouring |
| 8–11 | Both reach for the jar and the mug; more grounds pile up on the counter |
| 11–12 | The coffee-scooper stops and stares at the mess |
| 12–14 | Photo 2 pours milk where the mug just was, and it splashes across the counter |
How we made our first When Two Legends Try to Make Coffee AI video
We built it by hand first, the way you would without a template, so we know which parts matter.
1. We started from the trend clip
The trend comes from a TikTok post by @vibemotionai on 28 October 2026, captioned "When Two Legends Try to Make Coffee". The clip is about 14 seconds, so we kept all of it and set the length to 15 seconds: the milk at the end is the punchline, and cutting earlier would lose it.
2. We used one photo per person, in two different looks
Photo 1 was a red-haired girl in a red-trimmed school robe, photo 2 a blonde girl in a green-trimmed robe. Each image paired a face close-up with a full-body shot, so the model saw both the face and the whole outfit. Two very different looks also made it easy to check that nobody ended up in the other one's clothes.
3. We told the model who stands where
With two people in one kitchen, the model needs to know which face goes on which body. We named each role by its position and its prop: the one scooping coffee in front, the one with the steel kettle behind. We also asked for each outfit to come from its own photo. This is the prompt we ran, on Wan 3.0, 480p, 15 seconds, sound on, with the trend clip as the reference video:
Replace the two people in the video with the people from the reference photos. Each new person keeps their own face, hairstyle, skin tone and outfit from their reference photo. Keep every motion, camera angle, timing and the kitchen scene exactly the same. - The person scooping coffee (front right in the original) → the person in Image 1, wearing the outfit from Image 1. - The person holding the steel kettle (back left in the original) → the person in Image 2, wearing the outfit from Image 2. Vertical 9:16, handheld phone footage, natural indoor light, realistic faces, stable identity, no extra people, no text overlays.
4. It rendered in about 6 minutes
We submitted at 08:36 UTC and the video was ready at 08:42: about 6 minutes, first try, 300 credits. The file came out 14 seconds long. Both robes, both ties and both hairstyles stayed on the right person from the first scoop to the milk. That run is what the template now uses, with the prompt and settings already in place.
How to make a When Two Legends Try to Make Coffee AI video in 4 steps
Step 1: Open the template
Go to the When Two Legends Try to Make Coffee AI page. The template is already loaded at the top, with the model and prompt set up.
Step 2: Add the person who scoops the coffee
The first slot, Who scoops the coffee, is the one in front with the spoon. They're closest to the camera, so a sharp face matters most here. Knees up or full body works.
Step 3: Add the person who holds the kettle
The second slot, Who holds the kettle, stands behind and pours the milk at the very end, so they deliver the final mess. Pick whoever deserves that honor. No partner around? Click Use example under either slot and keep our sample photo.
Step 4: Generate and download
Press Generate. The button shows 300 credits before anything is charged. The video lands in My Videos, so you can close the tab. Download the MP4, which has no watermark, and post it.
Which photo plays which person in When Two Legends Try to Make Coffee?
| Slot | Role | Best photo |
|---|---|---|
| Photo 1 | In front, scoops the coffee onto the counter | One person, knees up or full body, face sharp |
| Photo 2 | Behind, holds the kettle and spills the milk | One person, knees up or full body, face clear |
To change who gets the milk moment, switch the two uploads.
What photos work best for a When Two Legends Try to Make Coffee AI video?
- One person per photo. A photo with both of you in it leaves the model guessing who scoops and who pours.
- Two different looks. Different hair and outfits keep each person in their own role and make the result easy to read.
- Show the outfit. Knees up or full body. A face close-up plus a full-body shot in one image worked well for us.
- Keep the face clear. No sunglasses, masks or hands over the face; the deadpan faces are the joke.
- Adults only. Use photos of yourself or adults who said yes. Photos of children, celebrities and public figures aren't allowed.
What to watch for in a When Two Legends Try to Make Coffee AI video
- The kitchen stays the same. The counter, the red mug, the kettle and the sound come with the template; faces, hair and clothes come from your photos.
- The reaction is small. Like the trend, nobody shouts. The coffee-scooper just stares at the mess around 11 seconds, and the milk does the rest.
- 480p is made for phones. It looks sharp on a vertical feed and soft if you blow it up on a big screen.
That's why we packaged it: the roles, the outfit line in the prompt and the model are already set, so you don't pay to rediscover them. Make your When Two Legends Try to Make Coffee AI video with the same settings.
Do you need the CapCut template for When Two Legends Try to Make Coffee?
No. CapCut template copies of the clip appeared from 29 October 2026, but they mean installing the app and editing it yourself. A photo template like ours runs in the browser: two photos, no filming, no editing, about $2.77 a video.
Is When Two Legends Try to Make Coffee AI free?
Not completely. New accounts get 75 free credits and one video takes 300, so you need the one-time $4.99 pack (540 credits). That covers one video at about $2.77 and leaves 240 credits, with no subscription. See pricing. Failed renders are refunded automatically.
How long is the video and how much does it cost?
About $2.77 for one video: 300 credits for a 14-second, 480p vertical clip with sound, ready in about 6 minutes.
What else is the trend called?
It's also called the zero synergy coffee making fail and the duo coffee trend; some posts call it two people try to make one coffee.
Can I make it with only one person?
Yes. Upload your photo in one slot and click Use example under the other, so our sample photo plays your partner.
Why don't we look like our photos?
Usually the photo is a group shot, a tiny face or has sunglasses. Use a sharp single-person photo facing the camera for each slot and run the template again.
Can I use photos of celebrities or football players?
No. Only photos of yourself or adults who agreed. Photos of celebrities, public figures and children aren't allowed.
Making another two-person clip? Our Hop Hop Hop AI Dance tutorial walks through a duo template with a switch in the middle.