The "meanwhile in their heads" video
Your two cats asleep on the sofa, then the caption "meanwhile in their heads" and the trap session: both shots edited back to back give the format going around on TikTok and Instagram.
Photos of your two cats become a live session: face to face at the hanging mic, they rap in English over a trap beat, sound included. See the workflow here, no account needed.
Real workflow: a grey and white cat and a black cat go into the same image, face to face at the hanging mic of a teal studio, then ten seconds of Seedance 2.5 with sound: a trap beat and one line rapped in English by each.
At a glance
Live sessions filmed against a wall of one single colour, mic hanging in the middle, are being remade right now with rapping cats, and those videos are everywhere. This workflow makes yours from photos of your two cats.
The hard part is keeping both your cats. Both photos go into the same image, which stands them face to face while keeping their fur, markings and eyes; the video then starts from that image without redrawing anything, and it is the video that adds the beat and the voices.
Seedance 2.5 renders the movement, the trap beat and both rapper voices in a single pass, in English, mouths in sync with the lyrics. The image and ten seconds of video cost two hundred and fifty-nine credits.
These are the files of the workflow above, exactly as the models produced them, unretouched. The starting point is shown when there is one.
Starting point
Render
Cat owners who post on TikTok and Instagram, creators who follow trends, shelters, pet brands and artists.
Two photos of your cats from the front, in daylight, plus one line of rap in English for each.
Give two sharp photos
Each cat from the front, in daylight, fur and eyes clearly visible. Two very similar cats get mixed up: if you have two tabbies, describe what sets them apart in the prompt.
Pick the wall colour
One saturated colour, wall and floor blended together: teal here, but orange, yellow or pink work too. That plain background makes the format recognisable at first glance.
Check the image before the video
Both coats, the mic between the two mouths, the tight framing: everything is judged on the nineteen-credit image, before paying for the video that starts from it as it is.
Write one line per cat
One short line in English between quotes for each, and the order they rap in. Ten seconds hold two lines of rap, not a whole verse.
The same canvas, tuned for different needs: each variation is two or three lines of prompt away.
Your two cats asleep on the sofa, then the caption "meanwhile in their heads" and the trap session: both shots edited back to back give the format going around on TikTok and Instagram.
One track per video and one wall colour per episode: your two cats become the recurring duo of an account, recognisable by their fur from one video to the next.
Photos of a friend's cats, and the video lands on their birthday: their name in a line of rap, ten seconds that land better than a message and get shared in the group.
Two cats up for adoption together in a trap session, their names in the lyrics: the video travels far more than an adoption listing and shows they are a duo.
A cat food or litter brand stages a customer's cats with their consent: the video looks like the trend of the moment, not like an ad, and makes people want to send in their own.
A rapper releasing a track has their cats rap the real line of the chorus: ten offbeat seconds that catch the eye and point to the full song.
Without the sentence that forbids it, the model falls back to a cat's cry instead of a rapper's voice. Keep "never meowing" and "real human rapper voices" in the prompt.
The model then guesses the coats, and the two rappers are just any cats. Photos from the front, in daylight, keep their markings and the colour of their eyes.
Ten seconds hold two lines of rap. A whole verse gets rattled off too fast or cut in the middle, and the mouth sync suffers.
The video starts from the image as it is: a cat that is too small or a badly placed mic stay there for ten seconds. Rerun the image, it costs far less.
Because it is the language of the clips going around, and English rap sounds right over a trap beat. The model raps the lyrics written between quotes: change them as you like, keeping the lines short.
Yes: delete the second Media node and rewrite the prompt for a single cat at the mic, rapping both lines. For a duet with you, see the duet with a giant cat.
Yes, replace "cat" with "dog" in both prompts. The image keeps the coat and head of the animals in the photos, and the prompt keeps the rapper voices.
Two hundred and fifty-nine credits: nineteen for the image on Nano Banana Pro, two hundred and forty for ten seconds of Seedance 2.5 in 720p with sound. A new image alone costs nineteen credits.
Measured on the demo next to Wan 3.0, from the same image: both rap the right lyrics over a real trap beat, but Wan 3.0 cut to a close-up halfway, whereas Seedance 2.5 holds the shot from start to finish, both cats at the mic. Wan 3.0 can be wired in instead at half the price.
Yes, up to thirty seconds, at the same price per second. Then write four lines and the order they are rapped in, so each cat gets its moment at the mic.
The template renders in vertical 9:16, full screen on a phone. For the horizontal version of the original clips, switch the image to 16:9: the video takes on its format.
Yes, these are your cats and music composed by the model. Label the content as AI generated when the network asks, as TikTok and Instagram offer to.
Make your cats rap over a trap beat with AI
Use this workflowFree account, no card required. The workflow opens pre-filled in your canvas.