Cinematic live-action zombie apocalypse scene, bright sunny daytime, Seoul, South Korea. A broken Han River bridge with a huge collapsed gap, twisted rebar and cracked concrete sticking out. A young Korean schoolgirl (long dark brown hair, wearing a pink Nike dress, white knee socks, sunglasses pushed up on her head) is being chased by a huge horde of zombies across the bridge. Shot 1: Wide shot, the girl leaps mid-air across the collapsed bridge gap over the river, hair flying, zombies at the broken edge behind her. Shot 2: Close-up, she barely grabs the rusty rebar at the edge and hangs over the dark river water, a terrified, desperate face, zombie hand reaching from above. Shot 3: She pulls herself up and sprints down the bridge road, a massive zombie crowd running after her, a white sedan and a green city bus on the road, Seoul apartment skyline in the background. Shot 4: She climbs onto the roof of the green bus while zombies pack the windows and reach out with their arms. Shot 5: Extreme close-up of her shocked, wide-eyed face, wind blowing her hair. Shot 6: She jumps across the tilted bus and wrecked cars at the second broken section, the zombie horde charging behind. Shot 7: Final dramatic leap over the river gap, two men reaching out from the edge, with a black military Black Hawk helicopter hovering at the bridge pier with a rescue rope hanging down. Style: photorealistic, high-octane action, handheld dynamic camera, slow-motion jumps, dramatic wide-angle and low-angle shots, realistic motion blur, sharp detail, cinematic color grading, blockbuster zombie movie look (Train to Busan style), 4K. Negative: cartoon, anime, blurry, distorted faces, extra limbs, text, watermark, low quality. Settings: 16:9, 20 seconds, Seedance 2.5.
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 · 0:20 · Cinematic · Horror · Dynamic. 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.