Produce a 12-second cinematic underground fashion short in a 16:9 aspect ratio, set at night inside an authentic multilevel concrete parking structure. Feature two fashionable adult women as the primary characters. The first model has long, sleek raven hair with straight-cut fringe, slim tinted sci-fi eyewear, a weathered metallic silver-and-charcoal sleeveless printed shirt, multiple steel-toned chains, layered wristbands, black half-finger gloves, relaxed dark pants, heavy black shoes, and an oversized aged leather-look shoulder tote. The other model has lengthy platinum-blonde hair beneath a snug dark knitted cap, large angular jet-black shades, a close-fitting ivory shirt, a roomy black coat hanging loosely below her shoulders, a cropped charcoal skirt, over-the-knee black stockings, thick-soled footwear, chrome accessories, and a spacious tactile purse. Film everything through an ultra-wide retro fisheye optic featuring obvious barrel warping and exaggerated spatial depth. The footage should resemble someone operating a compact first-generation digital camcorder with one extended arm. Retain minor compositional errors, gentle horizon wandering, organic hand tremors, tangible walking impacts, inconsistent distance from the models, authentic focus hunting, and intermittent directional blur. Avoid flawless stabilization, robotic tracking, or unnaturally polished camera motion. Begin inches away from both models as they squat beside a dark motorcycle and gaze coolly into the glass, with an overhead fluorescent tube creating a glowing halo above them. Swiftly circle, roll, and angle the camera around the pair while they reposition their eyewear, sweep back their hair, and playfully acknowledge the person filming. Jump to a striking ground-level macro view of the bike’s motor, back tire, chassis, shadowy steel surfaces, and polished chrome catching the severe ceiling illumination. Reveal both models kneeling beside an unfastened handbag, surrounded by convincing everyday belongings spread loosely across the cement floor. Add a momentary almost-black pulse synchronized with the rhythm. Resume from a close overhead fisheye viewpoint as they rummage through the purse. Have the fair-haired model lift a miniature metallic point-and-shoot camera directly toward the viewer. Let one model rise and step across the painted pedestrian crossing while her companion gathers the handbag. Trail them farther into the structure using relaxed handheld reverse tracking. Populate the environment with raw cement columns, sweeping vehicle ramps, mustard-yellow curb markings, bright directional symbols, a dark stationary sedan, and a pale cargo van positioned far behind them. Near the midpoint, flash a split-second three-frame editorial montage featuring the brunette from several intimate viewpoints, then instantly transition back into uninterrupted documentary-style footage. Capture her moving into the distance while carrying the weighty dark tote, followed by both models pivoting and striding assertively toward the operator. As they close the distance, let the curved optic organically magnify their facial features, fingers, accessories, and garments. Conclude beside the motorcycle as both women lower themselves near the camera once more, visually matching the first composition and forming an invisible continuous loop. Illuminate the sequence almost exclusively with functional cool-toned fluorescent fixtures found throughout the garage. Use a faint greenish-white cast, overexposed halos surrounding the lights, gentle yet dense shadows beneath the models’ eyes and garments, and restrained gray cement hues. Maintain tactile realism, imperfect handheld energy, raw editorial attitude, and an authentic early-2000s underground fashion aesthetic throughout.
This clip was generated with Seedance 2.5 by ByteDance. The prompt on this page is the exact text used to create it.
The example was generated with Seedance 2.5 using: 16:9 · 0:12 · Cyberpunk · Cinematic · Fashion. Reusing these settings gives the closest match.
Copy the prompt as-is first, then change the subject while keeping the camera, motion, and lighting terms. Video models are sensitive to wording, so expect a few runs.
Yes. Prompts work best as starting points — swap scenes, adjust pacing terms, or change the aspect ratio to fit your project.