AI video generation has changed the way creators bring ideas to life. With the right prompt, a simple concept can become a cinematic scene that looks like it came straight out of a movie.
Over time, I’ve tested different ideas and discovered that some prompts consistently produce more eye-catching and shareable results than others. In this article, I’m sharing 5 of my favorite cinematic AI video prompts that you can copy, customize, and use to create your own viral-style videos.
Whether you’re creating content for TikTok, YouTube Shorts, Instagram Reels, or simply experimenting with AI video generation, these prompts can help you turn ordinary ideas into something much more visually compelling.
Sky Diving
Video Prompt:
Ultra-realistic cinematic POV skydiving video, filmed with an action camera mounted on top of the second skydiver’s helmet. Scene starts inside a real skydiving airplane high above Dubai during daytime. The airplane interior shakes naturally with loud wind and engine vibrations. Two professional skydivers wearing authentic skydiving suits and parachute packs are standing near the open airplane door. The first skydiver looks directly into the camera with excitement and confidence, gives a clear thumbs up, realistic smile, then jumps out of the airplane into the sky. A second later, the camera wearer jumps after him, creating an intense dynamic freefall moment.
The camera follows the first skydiver closely through the air in a smooth but realistic action-camera movement. Sometimes the camera rotates naturally around him while falling, capturing his face, body movement, parachute gear, and the massive aerial view below. Show the stunning Dubai skyline beneath them, especially Palm Jumeirah clearly visible on the ground with realistic details and scale. The ocean surrounding Palm Jumeirah should look ultra realistic with sunlight reflections. A few soft distant clouds far away in the background, but the sky around the skydivers stays mostly clear and bright blue.
Extreme realism, cinematic action sports style, authentic skydiving physics, realistic body motion, realistic wind interaction on clothes and straps, natural lighting, dynamic camera movement, immersive freefall sensation, ultra high resolution, hyper detailed, breathtaking scale, adrenaline-filled atmosphere, realistic depth of field, motion blur only during fast movements, GoPro-style footage but cinematic quality, smooth transitions between close-up face shots and wide aerial views, epic and emotional energy, world-class action documentary style.
Nail Artist
Video Prompt:
Reference Image: Use the uploaded close-up photograph of the woman’s hand as the exact foundation of the video. Preserve the hand, skin tone, pores, fingerprints, wrinkles, lighting, perspective, and composition exactly as they appear in the reference image. The only visible transformation should be the gradual creation of a luxury manicure on one fingernail.
Prompt:
Create an ultra-photorealistic, cinematic macro video of a real woman’s hand where two tiny miniature characters, Remy and Alfredo Linguini from Ratatouille, are working together as professional nail artists. Their appearance must match the recognizable movie characters while being seamlessly integrated into a real-world environment with physically accurate lighting, shadows, reflections, and realistic textures. They should feel like they genuinely exist on top of the real human hand.
The entire video is one continuous shot with no cuts, no transitions, no time jumps, and no camera changes.
The camera is positioned as a professional 100mm macro lens in an extreme close-up. The framing stays almost completely locked throughout the sequence, with only imperceptible handheld micro-drift and subtle focus breathing typical of high-end macro cinematography. The movement should feel like a luxury cosmetic commercial.
Remy and Linguini are approximately 3–4 centimeters tall relative to the hand. Their scale feels believable and consistent throughout the video.
The woman keeps her hand relaxed. Very subtle natural breathing causes tiny finger movements. Whenever the finger moves, the miniature characters naturally react by adjusting their balance. Every footstep creates realistic contact with the skin, including tiny skin compression, believable weight transfer, soft contact shadows, and accurate friction. Nothing floats or slides unnaturally.
The sequence begins with Linguini carefully inspecting the fingernail while Remy opens a miniature professional manicure toolkit.
Remy hands tiny manicure tools to Linguini.
Linguini gently buffs the natural nail while microscopic nail dust becomes visible under the macro lens.
Remy prepares a transparent gel nail extension.
Linguini precisely aligns the transparent extension before pressing it naturally onto the nail surface.
Tiny air bubbles disappear as pressure is applied.
Remy hands him a miniature brush loaded with builder gel.
The builder gel behaves exactly like real cosmetic gel, slowly leveling itself with realistic viscosity, glossy reflections, and natural surface tension.
Next, Remy rolls in a tiny UV LED curing lamp mounted on miniature wheels.
The UV lamp emits a soft violet-blue glow that naturally illuminates the surrounding skin while curing the gel.
After curing, Linguini carefully applies an elegant nude-pink gel polish using slow, controlled brush strokes.
The polish spreads smoothly with realistic fluid behavior and self-leveling gloss.
Remy then hands over an ultra-fine detailing brush.
Working together, they create an elegant luxury nail design featuring delicate white floral artwork, subtle metallic gold accents, tiny crystal-like decorations, and premium salon-quality detailing.
The artwork appears handcrafted with incredible precision.
Linguini finishes by applying a crystal-clear ultra-glossy top coat.
The UV lamp cures the nail one final time.
The finished nail looks absolutely flawless with a premium luxury salon finish, mirror-like gloss, beautiful reflections, intricate artwork, and perfect edges.
Linguini smiles proudly.
Remy folds his arms with satisfaction while admiring their work.
Both characters sit naturally on the edge of the finished fingernail as the camera remains steady before the video ends.
The lighting is soft cinematic beauty lighting with premium cosmetic commercial aesthetics. Realistic global illumination, ambient occlusion, ray-traced reflections, subtle rim lighting, natural skin translucency, physically accurate shadows, and luxury product-commercial color grading create a refined, elegant look.
The skin must display visible pores, fingerprints, tiny hairs, realistic subsurface scattering, and natural imperfections. The nail tools must feature brushed metal textures with realistic reflections. Remy’s fur should display individual strands with natural movement. Linguini’s clothing should respond naturally to motion and gravity.
Every animation must follow real-world physics. Every object has believable weight. Every interaction creates proper contact. Every shadow updates dynamically. Every reflection behaves physically correctly. The hand remains photorealistic throughout.
The mood is calm, satisfying, highly detailed, elegant, luxurious, and relaxing, like a premium beauty advertisement combined with cinematic VFX.
Style: Ultra-photorealistic live-action + CGI integration, macro photography, luxury cosmetic commercial, cinematic realism, physically based rendering (PBR), ray-traced lighting, realistic physics simulation, global illumination, shallow depth of field, premium color grading, extremely high detail, film-quality rendering, 8K quality, clean image, no text, no subtitles, no logos, no watermark.
Spider Woman
Video Prompt:
SEEDANCE 2.0 — CINEMATIC LIVE-ACTION PROMPT
Duration: 15 Seconds
Aspect Ratio: 9:16 Vertical
STYLE
Ultra-realistic live-action, Hollywood blockbuster cinematography, AAA VFX quality, photorealistic, physically accurate lighting, premium CGI realism, natural skin tones, realistic cloth simulation, cinematic depth of field, volumetric lighting, subtle lens flare, realistic motion blur, seamless transitions, no cartoon look.
IMPORTANT
The first frame MUST perfectly match the uploaded reference image.
Preserve the character’s identity exactly:
same face, hairstyle, body proportions, clothing, colors, accessories, pose, expression, room layout, window position, lighting, and perspective.
The opening frame must be completely static and identical to the reference image before any movement begins.
────────────────────────
SCENE 1 — STATIC OPENING (0.0–4.0s)
The camera is completely locked on a tripod.
No camera shake.
No movement.
A young blonde woman with a short bob haircut sits on the edge of a high-rise apartment window overlooking Dubai during golden hour.
Behind her:
- Burj Khalifa
- modern glass skyscrapers
- warm orange-pink sunset
- atmospheric haze
- glowing skyline
She wears:
- navy cardigan
- white collared shirt
- navy tie
- grey sweater tied around her waist
- black shorts
- pink-and-white striped socks
- teal sneakers
- white fingerless gloves
Soft golden sunlight enters through the open window.
The room remains slightly lived-in with realistic furniture and subtle clutter.
The camera slowly performs an extremely gentle cinematic push-in.
She confidently looks directly into the camera and says:
“Alright… let’s do this one more time.”
She calmly places large over-ear headphones onto her ears.
She closes her eyes.
A slight smile appears.
Without hesitation she slowly leans backward and falls out of the open window.
────────────────────────
TRANSITION (4.0–4.1s)
The instant her body disappears outside the window, the entire screen becomes pure white for only 0.1 seconds.
This flash completely hides the scene transition.
────────────────────────
SCENE 2 — WEB SHOT (4.1–5.0s)
Immediately after the white flash, cut to an extreme cinematic close-up.
The camera is only inches from her right gloved hand.
Her fingers snap forward.
A highly detailed white web-line violently launches toward a nearby Dubai skyscraper.
The web stretches with realistic elasticity and physically accurate tension.
The sound and motion feel explosive.
The camera follows the web for a split second before being pulled into the swing.
────────────────────────
SCENE 3 — CONTINUOUS FPV SWING (5.0–12.0s)
Without any visible cut, transition into one continuous FPV drone-follow shot.
The camera remains only a few meters behind her throughout the entire sequence.
She swings between Dubai skyscrapers at incredible speed using consecutive web-lines.
Every swing has realistic momentum.
The ropes tighten naturally.
Her skirt, tie, hair, and clothing react accurately to the rushing wind.
The FPV camera dynamically:
- banks around towers
- dives between buildings
- flies above roads
- narrowly clears rooftops
- accelerates with cinematic speed ramps
- follows smooth arcs
- never loses her
The movement feels exactly like a blockbuster Spider-Man flying sequence.
She briefly:
runs across the reflective glass façade of a skyscraper,
pushes off,
drops toward the street,
passes above moving luxury cars,
crosses over palm-lined boulevards,
and immediately launches another web without slowing down.
No abrupt edits.
Everything feels like one seamless cinematic shot.
────────────────────────
SCENE 4 — FINAL HERO SHOT (12.0–15.0s)
She fires one final web.
It pulls her into an enormous upward swing above the entire Dubai skyline.
The FPV camera follows closely before smoothly pulling back into a breathtaking aerial reveal.
Burj Khalifa dominates the center of the skyline.
The sunset fills the sky with glowing orange, pink, and gold.
She reaches the highest point.
She releases the web.
She freely floats through the air.
She spreads both arms wide.
Laughing with pure exhilaration, she shouts:
“Woohoo!”
She rotates gracefully in slow motion while silhouetted against the glowing sunset.
The camera continues pulling upward and backward, revealing the entire city beneath her.
The final frame freezes on an iconic superhero pose suspended above Dubai.
────────────────────────
CAMERA LANGUAGE
Opening:
- Locked tripod
- Slow cinematic push-in
Transition:
- Perfect hidden cut through white flash
Action:
- Continuous FPV drone follow
- Smooth stabilization
- Dynamic banking
- Hero angles
- Speed ramps
- No visible cuts
Ending:
- Massive aerial pullback
- Epic skyline reveal
- Slow cinematic finish
────────────────────────
LIGHTING
Warm golden-hour sunlight
Soft atmospheric haze
Natural reflections on Dubai glass towers
Subtle anamorphic lens flare
HDR cinematic exposure
Volumetric sunlight
Hollywood color grading with warm highlights and slightly cool shadows
────────────────────────
QUALITY
Photorealistic live-action
Hollywood blockbuster quality
AAA visual effects
Physically accurate animation
Realistic facial expressions
Natural body mechanics
Realistic cloth and hair simulation
Seamless cinematic transitions
Ultra-detailed environment
No cartoon appearance
No low-quality CGI
No artifacts
No extra limbs
No duplicated body parts
No camera jitter
No frame glitches
No visible cuts
Prompt credit: miladramezani
Kids football
Video Prompt:
Ultra-realistic 8–10 second cinematic football match. Cute baby footballers playing in a packed stadium. A giant electronic stadium scoreboard is always visible in the top corner throughout the match.
Scene 1 (0–2s): Baby [Team 1] scores first. The crowd cheers. The scoreboard immediately updates from 0–0 to [Team 2] 0 – [Team 1] 1. The scoreboard stays visible for at least one second.
Scene 2 (2–5s): Baby Messi dribbles past three baby defenders and scores a beautiful equalizer. The scoreboard instantly changes to [Team 2] 1 – [Team 2] 1. Show the crowd celebrating and the updated scoreboard clearly.
Scene 3 (5–8s): Baby Messi receives a perfect pass, curls a spectacular winning shot into the top corner. The scoreboard immediately updates to [Team 2] 2 – [Team 1] 1. Hold the scoreboard on screen while fans celebrate.
Final shot (8–10s): Close-up of Baby Messi smiling while teammates lift him. Giant stadium screen and broadcast scoreboard clearly display FULL TIME — [Team 2] 2–1 [Team 1] with [Team 2] flags waving, fireworks, confetti, TV broadcast graphics, realistic football atmosphere, dramatic slow motion, 4K, broadcast-quality sports cinematography.
Extra instruction (very important):
The scoreboard must be clearly visible after every goal and remain visible in the final scene. Do not skip or change the score sequence: 0–1 → 1–1 → 2–1. Do not generate any other final score
Red Line
Video Prompt:
0.0–0.5 seconds | Opening to establish a positive tone
The shot starts from the most classic front-facing perspective of the concept art, fully revealing the group composition on the raft:
In the foreground lie recumbent and prone human figures, the middle section is packed with a throng of people, on the left stand an upright mast and a huge canvas, on the elevated right side stands a figure waving a piece of cloth, while the background consists of a stormy sea and a heavy, overcast sky.
At this moment, it feels as if the concept art has been directly “brought to life” into a 3D scene, yet the overall visual remains consistent with the original concept art.
The shot holds steady briefly, allowing the audience to clearly recognize that this is a front-facing composition of concept art.
0.5–1.5 seconds | Start advancing and orbit clockwise
The camera slowly pushes forward slightly, while starting to orbit the group of figures on the raft in a clockwise direction.
The camera glides smoothly along the front outer edge of the raft, gradually shifting from a front-facing perspective to a right-front angle.
At this stage, we need to make the audience feel that:
The corpses and recumbent figures in the foreground of the raft feature a clear sense of depth along the front-to-back axis
The crowded crowd in the medium shot and the figure with raised arms form a layered relationship.
The figure waving cloth at a height and the wooden barrels present a distinct spatial hierarchy.
The left mast and canvas begin to produce three-dimensional perspective changes towards the rear.
The waves rise and fall gently, with ripples lapping softly at the edge of the raft. The figure remains completely still in its original posture.
1.5–2.7 seconds | Circle to the right and rear to complete the main looping segment
The camera continues to orbit clockwise, passing the right side of the raft before moving into a rear perspective.
In this segment, the camera captures the following:
The volumetric relationship between the shirtless figure on the right and the prostrate figure
A standing figure at a height is observed from the rear side, creating a stronger sense of space
The overlapping layers of the central group sculpture on the raft are more distinct
The mast, canvas and the dark clouds behind form a huge three-dimensional structure.
Keep the camera at a low-to-medium position, slightly above the character’s chest to head on the raft; do not fly too high, and avoid excessive top-down perspective.
When circling, always stay around the entire group, do not deviate from the main subject, and do not go off to only look at the sea surface.
Fabrics, tattered sails, and flags are only allowed to sway very slightly; the seawater may surge gently; the clouds in the sky can have a faint sense of movement, but must not be turned into severe storm effects.
2.7–3.9 seconds | Move to the left rear and left side, then continue to complete the full lap
The camera continues to rotate clockwise, moving from the rear of the raft around to the left-rear side, and then to the left side.
This section focuses on demonstrating:
The huge mast and canvas on the left form a bulky mass from another perspective
The front-back occlusion relationship between the figures in the center of the raft
Figures lying on the ground, bending over, kneeling and raising their hands are interlaced and superimposed in the space
The relationship between the waves on the left and the edge of the raft
The three-dimensional effect of the damaged plank structure of the overall raft
Camera movement must remain smooth at all times: no sudden turns, no collisions, and no passing through canvas or human bodies.
All character movements shall remain completely unchanged; only the wind is allowed to gently stir the cloth strips and the edges of the sail, and the waves may nudge the raft gently but must not cause it to shake violently.
3.9–4.6 seconds | Return to the front perspective of the concept art
The camera continues to complete the final orbit, returning to the front angle consistent with the opening.
This moment should make the audience clearly feel that:
The camera has completely circled the raft group portrait and returned to the most iconic first-frame perspective of the original work.
At this point, the distribution of the characters, the position of the masts, the angle of the sails, the sea level, and the layers of dark clouds must all be identical to those at the beginning.
4.6–5.0 seconds | End with a slight zoom-out after returning to the first frame
After the camera returns to the front-facing position at the start, it does not hold the shot for too long, but instead performs a very subtle and elegant pull-back.
This pull-back shot should be very restrained; just gently pull the subject from the “front close-up of the concept art” to a slightly more complete and wider perspective as the ending.
Do not zoom out excessively, do not alter the frame composition, and do not make the raft appear too small.
The ending still retains the concept art tone, dramatic lighting and oppressive atmosphere.
