OpenZhenZhen AI Prompt Gallery · gpt-image-2.5 · image

Japanese Sweet Fashion Portrait in a Café

Japanese Sweet Fashion Portrait in a Café — explore this gpt-image-2.5 creation on OpenZhenZhen, with its original prompt and a reusable generation example.

Japanese Sweet Fashion Portrait in a Café — AI image creation
Japanese Sweet Fashion Portrait in a Café

Prompt

Chinese prompt

创作一张 9:16 竖版写实时装人像，主角是一位明确成年的 21 岁日本女性。她有小巧柔和的椭圆脸、自然略不对称且富有表现力的眼睛、柔和眉形、比例自然的鼻唇、淡淡的粉米色妆容，以及中长款柔和波浪中棕色头发。神情放松亲切，呈现甜美俏皮的日系辣妹风格，时髦自然而不夸张。保留真实肌肤纹理和可信的人体结构。 服装从来源中的环形细节和斜向垂坠获得灵感：黑色不对称时装上身，采用不透且稳妥的内衬，以小巧中央圆环点缀；搭配完全不透的深色裙装和轻薄网纱纱笼式外搭。领口与躯干保持完整覆盖；不得出现内衣、裸露内裤、透视躯干或性化的身体部位强调。整体应读作时装编辑造型，而非泳装或贴身衣物。 场景为休闲咖啡馆，使用符合室内环境的普通现有光线。呈现真实重力与布料物理效果、自然抓拍姿态和宁静日常氛围。画幅固定为 9:16 竖版。人物为成年人，并始终使用已配置的人物参考 ref-10a-v1。避免名人相似度、戏剧化情绪、塑料肌肤、解剖错误、意外暴露、文字、标志、水印或 UI。

English prompt

Create a realistic 9:16 vertical fashion portrait of a clearly adult 21-year-old Japanese woman. Give her a small softly oval face, naturally asymmetrical expressive eyes, gentle brows, a naturally proportioned nose and lips, subtle pink-beige makeup, and medium-long softly wavy medium-brown hair. Her expression is relaxed and approachable, with a sweet, playful gyaru-inspired style that feels contemporary and tasteful rather than exaggerated. Preserve natural skin texture and realistic anatomy. She wears a black asymmetric fashion outfit inspired by the source's ring detail and diagonal drape: an opaque, securely lined one-piece bodice with a small central ring accent, paired with a fully opaque dark skirt and a light mesh sarong-like overlay. Keep the neckline and torso fully covered; no lingerie, exposed underwear, sheer torso, or sexualized body-part emphasis. The look should read as editorial fashion, not swimwear or intimate apparel. Place her in a casual café with ordinary available interior light appropriate to the room. Use realistic gravity and fabric physics, a natural candid pose, and a calm everyday atmosphere. Keep the frame vertical 9:16. The subject is an adult and must use the configured human reference ref-10a-v1 consistently. Avoid celebrity likeness, dramatic emotion, artificial skin, anatomy errors, accidental exposure, text, logos, watermarks, or UI.

View API model

Explore more AI image and video prompts · Try image generation · Try video generation · Create on the AI canvas

  • The Moon Mender

    The Moon Mender — explore this Moss Image creation on OpenZhenZhen, with its original prompt and a reusable generation example.

  • Windblown Fashion Editorial

    Windblown Fashion Editorial — explore this Moss Image creation on OpenZhenZhen, with its original prompt and a reusable generation example.

  • Abstract Light Flame

    Abstract Light Flame — explore this Moss Image creation on OpenZhenZhen, with its original prompt and a reusable generation example.

  • The World Tree

    The World Tree — explore this Moss Image creation on OpenZhenZhen, with its original prompt and a reusable generation example.

  • Midnight Supercar Drift

    Midnight Supercar Drift — explore this Moss Image creation on OpenZhenZhen, with its original prompt and a reusable generation example.

  • Gothic Halloween Stage

    Gothic Halloween Stage — explore this Moss Image creation on OpenZhenZhen, with its original prompt and a reusable generation example.