A music video prompt is not a dance description. It's a production brief: which camera angle, which transition trigger, what the performer is doing at second three versus second nine, and what the light is doing when the beat drops. Generic prompts get generic footage. The prompts that generate something you'd actually cut into a real MV specify all of that — sometimes at the level of a frame-by-frame storyboard condensed into a single text block.
Here are 5 Seedance music video prompts from the Scenic gallery — covering vertical TikTok pop choreography, K-pop arena spectacle, power-duo stage timing, a single-shot dance film, and a mood-arc MV with multi-location wardrobe transitions. All free to copy.
1. The vertical pop MV — instant outfit changes, shot-by-shot breakdown
See the full prompt on scenic.sh →
"9:16 vertical music video, low-angle medium shot, a lively stylish young woman dancing directly to camera against a clean high-saturation sky backdrop, energetic pop commercial aesthetic, fixed camera for most shots, ultra smooth rhythm and choreography."
Why this works: With 89 likes, this is the highest-engagement music video prompt in the Scenic gallery, and the reason is structural: each shot is a numbered timestamp paired with an outfit, a gesture, an expression, and a transition cue. "Shot 1 (0-1s): cyan top and pale yellow skirt, hands on cheeks, points to camera, ends with a fast clockwise spin transition." "Shot 2 (1-5s): instant outfit change, finger-gun gestures, bright confident smile." There's no interpretive room left — the model doesn't have to invent the visual language, it just executes a production plan. The fixed camera instruction keeps everything readable against the high-saturation backdrop, and "colorful typography at top" cues overlay graphics that make each costume pop.
The takeaway: for vertical social music videos, break the clip into numbered shots with timestamps, outfit names, and explicit transition triggers (spin, swipe, flash). Each shot is effectively a mini-scene brief. The fixed camera is a constraint that forces performer energy to carry the whole frame — and it works because the gesture choreography is specified move by move.
2. The holographic concert spectacle — arena sweep to surreal twin reveal
See the full prompt on scenic.sh →
"0.0–2.5s: A breathtaking aerial sweep over a colossal concert arena. Thousands of fans pulse in sync, waving glowing light sticks like a galaxy of color. Massive LED panels ripple with animated visuals as lasers cut through mist-filled air."
Why this works: With 11 likes, this K-pop concert prompt earns its engagement by thinking at the scale of a stadium show — and by staging the prompt like a director's shot list. The structure escalates: wide establishing aerial (0–2.5s) → push-in to performer (2.5–5.5s) → reality-bending glitch effect (5.5–8.5s) → colossal hologram reveal (8.5–12s) → crowd reaction intercuts to final hero frame (12–15s). Each beat has a visual purpose: the glitch signals the genre-shift from documentary concert footage to fantasy; the hologram expands the performer's presence to arena scale; the crowd reaction validates the spectacle. The "identity consistency" block — "no variation in face, styling, or presence" — is the anti-drift instruction that keeps the central performer recognizable through all five beat shifts.
The takeaway: for K-pop or big-stage concert MVs, structure the prompt as an escalating five-act sequence (establish → push-in → genre-shift → spectacle reveal → crowd payoff). The holographic twin is a specific visual that Seedance handles well — but it only lands if you set up the scale with an aerial in the first beat.
3. The power duo — second-by-second timeline with character contrast
See the full prompt on scenic.sh →
"Ultra-realistic cinematic concert performance, high energy, bold charisma, dominant duo presence, like top-tier global pop idol stage. Hard-hitting, sharp, synchronized but with attitude contrast (male = strong/grounded, female = sharp/sassy)."
Why this works: With 10 likes, this duo performance prompt solves a hard problem: how do you keep two performers visually distinct while making them feel synchronized? The answer is character contrast encoded at the movement level. Every timestamp has two lines — male behavior and female behavior — describing the same beat with different energy. "Step touch right: male heavy grounded groove, female sharp lighter hit." "Hip roll: female smooth and expressive, male controlled and grounded." The contrast is written into the prompt, not left for the model to interpret. Stage design is specified as two-spotlight (red for female, white/blue for male), which maintains the distinction visually even when the performers share the frame. The second-by-second timeline (ten 1-second blocks) gives the model a movement grammar to execute.
The takeaway: for duo or group concert MVs, write character contrast into the movement description, not just the costumes. Separate lines per performer for each beat; specify which light source belongs to which character. The moment "they're different" is encoded in the action, not assumed from the context.
4. The single continuous shot — storyboard-to-dance-film with no cuts
See the full prompt on scenic.sh →
"Cinematic 15-second dance video, single continuous shot sequence, 24fps, 4K. Scene plays as one flowing continuous video: she stands still head down eyes closed → arms slowly rise → eyes snap open extreme closeup → head whips sideways hair flying → explosive arm wave full body."
Why this works: With 6 likes, this prompt is a masterclass in writing for continuous camera rather than for edited sequence. Every movement in the chain is connected — there are no cut points, so each phrase flows into the next via specific bridging motion: "spinning fast outfit blurring → sudden sharp freeze pose." The pacing variation is built into the prompt itself: motion blur on fast movements, slow motion on the landing and close-up moments. A 21-panel storyboard image (used as a reference) provided the skeleton; the prompt converted it into a unified continuous flow with amber chiaroscuro lighting holding visual consistency across all twenty-one beats. "The word DANCE appears" at the end functions as a narrative landing: the physical performance resolves into an identity statement.
The takeaway: when shooting for a single-take effect, chain each movement into the next with a bridging action, rather than listing standalone beats. Use motion blur / slow motion modifiers within the same prompt to create pace variation without editorial cuts. A storyboard reference image dramatically improves motion fidelity for complex continuous sequences.
5. The mood-arc MV — wardrobe transitions across six environments
See the full prompt on scenic.sh →
"High-energy modern pop girl group music video, ultra-polished cinematic style, glossy lighting, sharp contrast, beat-synced editing, luxury idol aesthetic, 10 seconds, 140 BPM. Wardrobe Transitions: Red-black street performance outfit → Elegant black-gold shimmer → Futuristic metallic/chrome → Edgy all-black asymmetrical → Final iconic red + gold accents."
Why this works: With 7 likes, this prompt is built around mood escalation, not just choreography. The five-outfit arc — street performance → shimmer elegance → chrome futurism → edgy black → iconic finale — is a complete emotional journey embedded in the wardrobe. Each environment (neon alley, pastel dance studio, LED tunnel, rooftop, reflective stage) maps to one mood, and "transitions triggered by hand gestures, spins, lighting shifts, shadow morphs" is the instruction that makes the wardrobe changes feel cinematic rather than jarring. "No glitch effects" rules out the cheap transition — it forces the model to use organic motion-based cuts. The BPM spec (140 BPM) tells the model what editing tempo to emulate, which sharpens the snap cuts and beat-sync hits without over-specifying each individual moment.
The takeaway: for a mood-arc MV where the performer evolves across the video, write the wardrobe and environment sequence as a paired list — each outfit maps to one location and one emotional register. Specify the transition mechanism (spin, shadow morph) rather than the transition style; a BPM cue governs the editing tempo without requiring a shot-by-shot list.
Music video prompt cheat sheet
Across all five, the structural techniques are consistent and portable:
- Shot-by-shot breakdown — numbered timestamps with outfit, gesture, and transition trigger; the model executes a production plan rather than improvising.
- Escalating five-act sequence — establish scale → push in → genre-shift → spectacle reveal → crowd payoff; each beat has a visual purpose.
- Character contrast in movement — for duo/group work, write separate behavior lines per performer on the same beat; contrast is encoded in action, not appearance.
- Continuous chain — link each movement to the next with a bridging action; specify motion blur vs. slow motion inline to pace the flow without editorial cuts.
- Wardrobe-environment pairing — map each outfit to a location and emotional register; a BPM cue governs editing tempo without a full shot list.
Browse more dance and performance prompts in the Scenic action scenes gallery, or read how to write Seedance 2 prompts to build your own from scratch.
