How does a couple coffee AI video work?
A couple coffee AI video turns two solo photos into one clip in which both people stand in the same kitchen and ruin the coffee together. Each photo supplies one face, one hairstyle and one outfit. Upload yours.
From two photos to one clip
You are not editing two videos together. The template already holds the kitchen, the movement and the timing, and it has two empty places in it. Your first photo fills the front place, where the spoon is. Your second photo fills the back place, where the kettle and the milk are. The AI video model, Seedance 2.0 / 2.5 by ByteDance, then draws both people into the scene. The result is a 14–15-second MP4 with the trend sound on it. See what the AI part means here.
Photo tips
- Use a separate photo for each person. A cropped group shot makes the faces harder to separate.
- Choose a front-facing, evenly lit photo with the face, hair and shoulders in view.
- Wear what you want to see. Outfits are copied from the photo.
- Skip sunglasses, heavy beauty filters and tight crops.
- Expect small drift. Faces and hands can change between frames, and the person further back is the hardest to keep.
Consent comes first
Both people must be adults and both must say yes before a photo is uploaded. Do not use a photo of an ex, a coworker or anyone who has not agreed, and do not use a celebrity or a public figure. Never make a video that is meant to fool people into thinking it is real. If someone asks you to remove a video of them, take it down. Our acceptable use rules set out the limits.
Roles and the next step
Who holds the spoon and who pours is covered on the couple coffee page. A render usually takes a few minutes, and failed renders return their credits.