sup.video analyzes your song down to every kick, snare, and section change — then gives you a professional timeline to direct AI-generated visuals that actually move with the music.
Drop any track. We handle the audio analysis — stems, beats, lyrics, structure.
A beat-synced, multi-scene music video with transitions timed to your song's energy.
The same audio intelligence and timeline tools adapt to whatever you make. Describe the visual world — sup.video builds it.
"Rainy window, soft focus, girl sketching in a warm apartment"
"Street-lit night drive, low-rider reflections, smoke and city glow"
"Golden hour meadow, 16mm film grain, bare feet on dirt road"
"Laser grid tunnel, chrome figures dancing, strobe-cut edits on every snare"
Storyboard
4 scenes
Neon alley walk

Close-up portrait

City panorama

Club bass drop
Master Prompt
Video Model
Pipeline
Hover over any element to explore the editor
Swipe to explore the timeline
Describe your video in one prompt. AI generates characters, reference images, and scenes — each mapped to verse, chorus, or drop.
Waveform analysis detects every beat. Energy curves track intensity. Song structure (verse/chorus/drop) is mapped automatically.
AI-generated video clips, beat-synced transitions, and word-level lyric overlays — all on a professional editing timeline you control.
Isolated kick drum drives bounce, shake, and zoom effects. Toggle per clip and fine-tune intensity with a slider.
Pick the best AI model per clip — Kling 3, Seedance, or Wan. Watch every processing step happen in real-time.
Describe your vision in one sentence. Our AI planner analyzes your song structure, extracts characters, generates reference images, and maps scenes to verse/chorus/bridge — all automatically.






Select a time range, preview start and end frames, describe your transition. Generate up to 4 options per transition using Kling 3, Seedance, or Wan — pick the one that hits right on the beat.
We isolate the exact kick drum waveform and drive visual effects from it. Not just on-beat — the actual shape of each kick controls the bounce. A punchy 808 feels different from a soft house kick.

Drop a song and sup.video instantly separates drums, vocals, bass, and melody. Every kick hit, snare, and downbeat is detected. Lyrics are transcribed with word-level timing. Song structure is mapped automatically.
Select any storyboard image, paint over the area you want to change, and describe what to put there. Powered by FLUX Fill — seamless, context-aware inpainting that preserves the rest of your image.

Route each clip to the best AI video model. Kling 3 for action, Seedance for rhythm, Wan for style. Switch models per-clip on the timeline — no lock-in.
4K 60fps, cinematic action, up to 2 min
Social-optimized, rhythm-aware, audio sync
Stylized, artistic, strong character consistency
Free to start. No credit card, no production crew required.
Make your first video free