Original document
GPT Image 2 · en
A stylish anime illustration depicts a young woman standing on a modern city street in the rain, shot from a high-contrast, low-angle perspective. She has {argument name="hair color" default="bright pink"} Her hair was tied in a messy low bun, with a few stray strands falling on either side of her face. She had fair skin and cool, slightly aloof light gray-green eyes. She wore a black baseball cap with elegant white handwriting on it {argument name="cap text" default="Butterflies"} She wore several small earrings and dangling gold star-shaped earrings. A thin stick, like a lollipop stick or candy stick, with a small star decoration at the end, was casually held in her mouth, her hand gently touching it. Her outfit was street-style chic, shimmering slightly in the rain: a black cropped vest, an oversized, semi-transparent white jacket slipping from her shoulders, one sleeve adorned with delicate gold patterns, layered silver necklaces, and a dark-colored wristwatch. In her right hand, she held a transparent umbrella, its surface reflecting the wet light and raindrops. To the right of the background was a massive, curved glass building, below which lay a blurred cityscape. After the rain, a bright blue sky revealed itself, and a vibrant rainbow arced across the upper right sky. The emphasis was on the shimmering, wet surfaces, the refracted light, the glistening raindrops, the reflective glass, and the slightly bright iridescent halo. Highly detailed comic-style lines, soft painterly shadows, bright highlights, cinematic composition, stylish urban atmosphere, and exquisite and luxurious texture.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.
