
A cinematic photorealistic cyberpunk action scene in heavy rain at night. A dark-haired woman in a long black leather trench coat stands on a narrow elevated industrial walkway surrounded by neon city lights and wet reflections. She is performing a dramatic bullet-time dodge, leaning her upper body backward while keeping both feet on the ground. **Every bullet travels in exactly the same direction: from the LEFT side of the image toward the RIGHT side of the image.** **Every bullet tip points RIGHT.** **Every bullet trail extends behind the bullet toward the LEFT.** The bullets pass horizontally across the frame at different heights. One bullet passes just above her chest, one passes directly in front of her face, one narrowly misses her shoulder, and another passes near her waist. None of the bullets point toward the camera or away from the camera. None point toward her from different directions. No circular, radial, halo, orbiting, or symmetrical bullet arrangement. The bullets are caught in extreme slow motion while cutting through the heavy rain. Each bullet leaves a narrow wake of displaced water droplets and mist directly behind it. Tiny suspended rain droplets make the bullet trajectories clearly visible. Her wet hair streams toward the RIGHT from the sudden movement and wind. Her soaked trench coat flows naturally behind her body. Her expression is focused and tense as she narrowly avoids the incoming volley. Full-body side-view composition, camera perpendicular to the bullet trajectories, clearly showing the woman and the bullets in profile. Strong teal-blue backlighting, wet metal railings, reflective puddles, rain haze, distant neon signs, cinematic depth of field, realistic anatomy, physically coherent trajectories, detailed water interaction, dramatic high-end science-fiction action film still. **Important composition rule:** the image must read instantly as several bullets traveling left-to-right while a woman bends backward to dodge them.
Parameters used to generate this content
The node graph used to generate this content — pan and zoom to explore, or download it to run in ComfyUI.
AI models used to generate this content