File, frame and length
A vertical 9:16 MP4 of 5 to 15 seconds, from My videos. Square and landscape variants are one click away and never change the price.
AI music video ยท song in one render
Add one or two photos and a line about the occasion. The generator writes the rap, builds the beat, then films the people in your photos performing it.
What it looks like






The short answer
An AI music video generator makes the song and the video in one render. Give it a photo of each performer and a line about the occasion; it writes the rap, builds the beat and films those people performing it.
A standard render (720p, 12 seconds, vertical) costs 10 credits, $6.99 one-off, and a failed render is refunded. Full pricing and credit packs have their own page.
Three moves
Front-facing and evenly lit, one person per photo. In duo mode the left photo is the lead and the right one trades bars; one-photo mode makes a solo. They never have to be shot together.
Why: the performance is rebuilt from the face, so a clear face beats a stylish crop.
Four stages, four models, three resolutions, five lengths, five ratios. The default is 720p, 12 seconds and 9:16 โ the cheapest setting that looks right on a phone.
Why: the stage sets the picture and the length sets the price.
Type the names and one detail โ โJake turns 30 and still can't cookโ โ pick an occasion and press create. 200 characters; empty gives a generic verse.
Why: a real name gives the lyric writer something to rhyme around.
The output, item by item
A vertical 9:16 MP4 of 5 to 15 seconds, from My videos. Square and landscape variants are one click away and never change the price.
720p is the default: the cheapest render that holds up full-screen on a phone. 480p roughly halves the credits, 1080p needs a Pro model.
In duo mode the left photo takes the mic and the verse while the right one hypes and answers. Solo mode hands the verse to one performer. The left slot decides it, not a toggle.
The model writes original bars and an original beat around your line. No pre-made tracks, and no famous voice is imitated.
The orange lobby reads as the Hotel Lobby trend, the marble lobby as a formal occasion, the purple booth as studio footage, the night street as a battle. The stage never changes the price.
Mouth shapes are generated against the vocal rather than pasted from a template, so an open-mouth photo lands better than a closed smile.
10 credits at 720p and 12 seconds, or $6.99 one-off. Longer or sharper settings scale with the rate on each model card.
What people make
โJordan turns 30 and still can't cook pastaโ
The lyric names the person and the age, and the guest of honour performs it with you. Play it at the table or post it in the chat.
โTen years of us, and the first date that went wrongโ
Both of you in one frame. The verse counts the years and keeps the detail you gave it, which makes it yours rather than a template.
โBest friends since second grade, still undefeated at everythingโ
Two photos and one shared memory. It gets reposted with the sound on, because the joke is inside knowledge.
โRosa's bakery, sourdough that sells out by nineโ
Name the brand and what it does, then let the verse carry the shouting. Good for a launch, a sale or a store anniversary.
โMochi the corgi runs this house and everyone knows itโ
A clear photo is all a performer needs, so a dog, a cat or a pair can carry the verse. The photo in your camera roll is the whole input.
โPriya is leaving, and she was the only one who fixed the printerโ
A leaving gift that costs less than a signed card and gets watched more than once. The colleague performs and the joke is the topic.
What good looks like
Front-facing, evenly lit, one person per photo, 300 px minimum. Sunglasses, hard shadow and group shots are the usual reasons a render misses the person.
A verse about a birthday, a new job and a dog becomes a verse about nothing. Pick one occasion, one person and the detail they would recognise.
The orange booth reads as trend, the street cypher as confrontation, the studio booth as process. Choose the reference your viewers know.
Short enough to loop without a pause, long enough for a hook and eight bars. A second scene is a second render: cut the clips in your editor.
Before you pay
What it charges, what comes back, and what you can post.
Add a photo per performer, pick a stage and an occasion, type a line about it, then press create. The song and the video come out of the same render.
No free preview: rendering costs compute, so it is paid from the first video. A standard render is 10 credits, or $6.99 one-off. Credits never expire and a failed render is refunded.
Not on this page: the generator writes and performs an original rap. For visuals on a recording you already have, use the MP3-to-video generator, which keeps your audio.
No filming, no green screen, no editor. It runs in the browser and returns an MP4. An account is required, because the render is tied to your credits and saved to My videos.
An MP4 with no watermark, in 9:16, 1:1, 16:9, 3:4 or 4:3, at 480p, 720p or 1080p depending on the model. Vertical is the default because these clips are watched on a phone.
Yes. The MP4 has no watermark and lands in My videos, ready to post.
About a minute for a standard 12-second render; a busy queue or a Pro model takes longer. The task keeps running if you close the tab and reappears in My videos.
The stage is set
One or two photos, one line about who it is for, one render. Change the settings after the first result.
A render that fails is never charged. Credits return on their own.