Visualizers and lyric backgrounds
Moving backgrounds for lyric videos and visualizers, in 16:9 for YouTube or 9:16 for Shorts and Reels.
EveryGen AI Music Video Generator
Describe the look of your song and attach a short excerpt of the track. EveryGen AI renders shots that take their mood and rhythm from it, up to 30 seconds each.
Prompts to start from
What is the EveryGen AI music video generator?
It is the EveryGen AI video agent set up for music videos. You describe the scenes, the artist or the visual style, and the agent writes the shot prompts, picks a model and renders every clip with sound in the same chat.
Each render runs 4 to 30 seconds, so a music video is built shot by shot: an opening, the verses, the chorus, a closing image. Attach an audio excerpt (MP3, M4A, WAV or OGG, up to 100 MB) as a reference for mood and rhythm, then download every shot as MP4 and cut them together in your editor.
What musicians and creators make with it
Story music videos
Plan the song as scenes and render them one by one, repeating the same character description and reference image so the cast stays recognisable.
Performance shots
Turn a photo of the artist into a moving performance shot, with the photo as the opening frame and an optional closing frame.
Release teasers
Short vertical clips to announce a single or a release date on TikTok, Reels and Shorts.
How to make a music video with EveryGen AI
- 1
Describe the song
Tell the agent the genre, mood, tempo and the scenes you picture. Attach an excerpt of the track or a photo of the artist if you have one.
- 2
Render the shots
The agent writes a prompt for each shot and renders it with Seedance 2.5 by default, up to 30 seconds a clip. Ask for 16:9 or 9:16 depending on where the video will run.
- 3
Revise and assemble
Reply with changes for any shot, then download the MP4s and line them up against the full track in your video editor.
EveryGen AI music video specs
| Input | A written prompt in any language. Images, clips or audio can be added as references. |
|---|---|
| Uploads | JPG, PNG and WEBP images, MP4, MOV and WEBM clips, MP3, M4A, WAV and OGG audio. Up to 100 MB per file and 10 files at a time. |
| Models | MiniMax H3 and Seedance 2.5 |
| Clip length | MiniMax H3: 5–15 seconds; Seedance 2.5: 4–30 seconds |
| Resolution | MiniMax H3: 768P / 2K; Seedance 2.5: 480p / 720p |
| Aspect ratios | 21:9, 16:9, 4:3, 1:1, 3:4, 9:16, or picked automatically |
| Sound | Generated together with the picture on every model |
| Cost | MiniMax H3 at 2K: 100 credits per 5 seconds; Seedance 2.5 at 720p: 400 credits per 5 seconds |
| Output | MP4 video with audio, saved to the chat and your library |
Music video prompts to try
Click a prompt to load it into the composer, then edit it before you send.
Synthwave night drive
“A night drive along a coastal highway in a vintage convertible, neon pink and cyan city lights reflected on the wet road, slow dolly alongside the car, synthwave mood, 16:9.”
Acoustic ballad
“A singer with an acoustic guitar on a wooden porch at golden hour, warm backlight through floating dust, slow push-in, intimate and calm.”
Rooftop dance
“A dancer performing sharp moves on a city rooftop at dusk, the skyline behind, the camera circling low, punchy rhythm, vertical 9:16.”
Dreamy visualizer
“Slow-motion ink clouds in deep blue and gold blooming in water, abstract and dreamy, gentle camera drift, a background visual for a lyric video.”
Tips for better AI music videos
- 1
Name the genre and the tempo in the prompt: "slow dreamy ballad" and "fast punchy drill beat" lead to very different camera movement.
- 2
Plan one shot per section of the song and keep each one under 30 seconds; a three-minute track needs several shots.
- 3
Repeat the same character and style description in every shot, and reuse the same reference image, so the video reads as one piece.
- 4
Use only music you own or have the rights to, and leave room in the edit to cut on the beat.
EveryGen AI music video generator: FAQ
Not in a single render. Each clip is 4 to 30 seconds, so you build the video shot by shot in one chat and assemble the downloaded MP4s in a video editor.
Yes. Attach an excerpt as MP3, M4A, WAV or OGG, up to 100 MB per file. The agent passes it to the model as an audio reference for mood and rhythm. Seedance 2.5 can use audio on its own; MiniMax H3 needs a reference image or video with it.
Lip-sync to an uploaded track is not guaranteed. Performance shots, dancing, instruments and abstract visuals come out more reliably than close-up singing.
Seedance 2.5 is preselected because it renders the longest takes, up to 30 seconds, and can follow an audio reference on its own. Switch to MiniMax H3 when you need a sharper 2K picture.
Each shot is priced by model, length and resolution; a 5-second MiniMax H3 clip at 2K costs 100 credits. The composer shows the price before you send, and the pricing page lists every rate.
EveryGen AI is an independent product. Seedance is a trademark of ByteDance and MiniMax is a trademark of MiniMax; EveryGen AI is not affiliated with either.
More from EveryGen AI
EveryGen AI Video GeneratorStart on the homepage composer with every model and mode.EveryGen AI Image to VideoUpload a photo as the opening frame and animate it with sound.EveryGen AI Text to VideoType a sentence and get a clip with sound.EveryGen StudioThe chat workspace where every project, take and upload lives.EveryGen AI Seedance 2.5The longest takes EveryGen AI offers, with soundEveryGen AI MiniMax H32K video with sound at a lower cost per clipEveryGen AI Video Ad GeneratorProduct videos, commercials and UGC-style adsEveryGen AI Animation GeneratorCartoon, anime and stop-motion clips from text or drawingsEveryGen AI Prompt LibraryReal prompts next to the clips they made, ready to load.EveryGen AI PricingPlans, credits per clip and how refunds work.
Start your music video in EveryGen AI
Describe the song, attach the track and render the first shot.