Original document
GPT Image 2 · en
A candid street style portrait shows two adorable anime-style girls standing side-by-side in front of a colorful graffiti wall on a sidewalk in a Japanese city. The photo was taken at eye level, creating a shallow depth of field and soft, overcast natural light. Both girls have fair, smooth skin and short, straight bob haircuts, with bangs partially obscuring their foreheads. Their faces are deliberately simplified or blurred, giving them an anonymous, doll-like appearance. The girl on the left has {argument name="hair color" default="silver-gray"} The girl on the right has icy blue hair, wears a black knit hat with a rectangular "Nightfall" logo patch, a black leather jacket over a white printed T-shirt (with dark distressed illustrations and purple accents), layered silver chain necklaces (including heart-shaped pendants), a black choker, and carries a black backpack on one shoulder. The girl on the right has icy blue hair, wears a black knit hat with a white graffiti-style logo, an oversized black hoodie with striking bright pink Japanese graffiti lettering on the chest, a silver chain necklace, and carries a black backpack. They are relaxed and stylish, leaning slightly towards each other, one hand lightly touching the shoulder of the hoodie. The background is a brightly colored pink, blue, and purple graffiti wall, with a telephone pole covered in stickers on the right edge, including layers of street stickers such as "SMILE," "AQUARIUS," and a red sticker that reads "HELLO my name is," covered in graffiti. It emphasizes Japanese street style, alternative fashion, cute and cool atmosphere, realistic fabric texture, clear clothing details, subtle cinematic color grading, and trendy Harajuku snapshot aesthetics.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.
