Reference example · not your generated result
Make your coffee-fail video in three steps
- Choose two separate photos, with one adult person or one pet in each. Put the first character in the left-hand slot and the second in the right-hand slot. Use clear JPG, PNG, or WebP images, each no larger than 30 MB; check the two previews before continuing.
- Sign in and create your video for 100 site credits. If you leave the page to sign in, select your photos again when you return. The kitchen, coffee-spilling action, and soundtrack are already set; there is no reference clip to find or instruction box to fill in.
- Watch the finished 15-second, 9:16, 720p video with sound before downloading the MP4. Check both identities and the spill sequence. Your library is the place to revisit the creation; checking an existing task does not start another generation. A new generation costs another 100 credits.
What is the Coffee Making AI trend?
The coffee-making joke is a small disaster played completely straight. Two characters stand at a kitchen counter, apparently ready to make an ordinary drink in a mug. Then the coffee goes outside the cup, the milk spreads across the worktop, and the performers seem more interested in finishing their routine than dealing with the mess. The contrast between their serious expressions and the increasingly hopeless preparation gives the clip its punchline. You do not need an elaborate caption for viewers to understand what has gone wrong.
A dog version is associated with @maxboerboeldog, and there are also cat, fantasy-character, and human remakes. This page follows the messy kitchen format, not a different couple trend about calmly drinking coffee together. The source links below let you compare that distinction yourself. We do not claim to know the first-ever upload or how quickly the trend is growing. What matters for your remake is the action: a shared counter, two recognizable characters, an escalating spill, and a cup that somehow remains the priority.
This template supplies the scene and movement so you can concentrate on casting the double act. Your two photos supply the identities, rather than a description of an imaginary barista. The generated result is fictional entertainment, not footage of your pet actually handling a kettle or drinking coffee. Keep that distinction clear if you share the video with people who have not seen the format before.
Choose two photos that make the double act readable
Use one subject in each image, with the face easy to recognize. A well-lit portrait is more useful than an impressive but shadowy shot where sunglasses, hair, or another person hide the features. For adults, choose a reasonably front-facing view with some shoulders or upper body visible. For animals, include the head, ears, and enough of the body to show the coat and silhouette. You do not need to stage a real kitchen photo or hold a cup before uploading.
The upload slots decide the roles: photo one supplies the character on the left, while photo two supplies the character on the right. Look at both previews before you press create. If you picked the wrong file, remove it from that slot and choose another. When deciding between two portraits, compare the eyes and expression rather than the background. A face looking towards the camera is easier to assess than one hidden behind a raised cup or turned away from the lens.
A desktop visitor can also use the phone-upload control next to each slot. Transfer one image into the intended role, then repeat for the other role. Receiving a photo only fills the preview; it does not create a video or spend credits. Whether you select files locally or send them from a phone, the same photo checks apply. Keep sensitive information out of the background and use images you have permission to upload.
Let your dog or cat run the coffee counter
Pets are a natural fit for the straight-faced coffee disaster. A dignified cat becomes funny when the counter turns into a milk puddle, while an enthusiastic dog makes a convincing assistant who has completely misunderstood the task. You can pair two pets, choose two adults, or put an adult and a pet together. Both slots still need a photo: this page does not offer a one-upload mode that silently chooses a partner for you.
For a cat, pick an image with visible eyes and clear ears rather than a tiny face buried in a blanket. For a dog, look for an expression and coat markings that remain readable on a phone screen. If the pair look similar, distinctive collars or different fur patterns can help you tell them apart, but extra accessories are not required. The two photos need not have been taken at the same place or time, because the coffee-counter setting comes from the template.
The result animates animals in an intentionally impossible role. It can also introduce odd paws, changing markings, or an expression that does not quite resemble your pet. Review those details before sharing. Never recreate the scene with real animals, hot liquids, or coffee; the appeal is that the entire kitchen performance is imagined. A recognizable, uncluttered source photo gives you a better starting point, not a promise of perfect identity preservation.
A couple coffee fail without writing a prompt
For a partner or friend version, choose the two expressions before you worry about a caption. A composed face beside someone who looks permanently amused can make the reaction feel like an inside joke. Use consenting adults and avoid casting a person in a scene they would find embarrassing. The aim is a shared comedy, not a deceptive video that makes someone appear to have done something in real life.
There is no need to describe every spoonful, upload a reference video, or decide which camera should move. The template carries the kitchen setup and the progression from coffee grounds to spilled milk. You choose the two identities and confirm the credit cost. That also means this page is not a general-purpose coffee-film editor: there are no settings for swapping the action, removing one character, choosing a different duration, or rewriting the scene.
If you want a completely different story, the related text-to-video page is a more suitable starting point. For this coffee fail, watch how both characters behave in the final result. The comic timing should still make sense without a caption explaining every movement. AI can change facial detail or the way an arm meets a cup, so treat the result as a new interpretation of the scene rather than an exact recording of the people in your photos.
Think about where the clip will appear before choosing a caption. In a private conversation, a short line about who should never be allowed near the coffee machine may be enough. For a wider audience, make the fictional setup clear and let viewers see the two characters before the biggest spill. Watch once without looking at your caption: can you follow the cup, the coffee grounds, and the milk? If the action is difficult to read, a longer explanation will not necessarily make the video funnier. Review the faces and the ending as well as the first frame, and share only a version you are comfortable showing to the people pictured.