Original document
GPT Image 2 · en
As dusk falls, a cinematic, hyper-realistic action scene unfolds on the rain-soaked city streets, showcasing a sudden multi-vehicle collision and explosion in the heart of a modern city. In the foreground, an adult male with an indistinct face, wearing a dark jacket and trousers, runs erratically and unsteadily across the road towards the camera, seemingly fleeing danger. To his left, a dark gray sedan with its headlights on is splashing water from the road. In the mid-ground on the right, two wrecked cars are colliding: a silver sedan with its driver's side door open, and a dark-colored vehicle flipping in mid-air, surrounded by shattered glass, torn metal, and flying debris. In the background, two fireballs are visible: one in the distance on the left side of the street, and the other a large explosion on the right, both spewing sparks and black smoke into the sky. The image shows thick smoke rising above the dense, slightly out-of-focus high-rise buildings of the city center. It includes wet, reflective asphalt pavement, scattered car parts, glowing streetlights, utility poles, overhead power lines, and a melancholic, blue-gray storm atmosphere with orange flames reflected in puddles. The composition should use a wide-angle, low-angle shot, emphasizing strong dynamism, dramatic perspective, cinematic depth of field, realistic physics, volumetric smoke, swirling embers, and highly detailed disaster film lighting. Add a white [object/feature] in the upper right corner.{argument name="watermark text" default="Pollo.ai"} Watermark.How to use
- Copy the complete original prompt, including any parameters.
- Replace the subject, text and composition with your own requirements.
- Use a compatible external tool; check its current access conditions and review the result.
Before you start
- Check whether a reference image is required. A compatible model label does not mean this example was generated with that model.
- Keep names, figures and visible text accurate; results can vary between model versions.
