OpenZhenZhen AI 提示词画廊 · gpt-image-2.5 · image

六条指定短语的繁华街头成年风衣肖像

Portrait of an adult in a trench coat on a bustling street with six specified phrases

绑定身份参考的成年模特身着米色风衣，在繁忙城市街头回眸，呈现黄色出租车光斑与六条清晰指定短语的4:5长焦肖像。

六条指定短语的繁华街头成年风衣肖像 — AI image creation
六条指定短语的繁华街头成年风衣肖像

提示词 / Prompt

中文提示词

使用配置身份参考ref-10a-v1作为明确成年模特的身份参考。创作一张原创超写实的4:5竖幅街头时尚肖像，场景为繁忙的普通城市街道。将模特略偏左居中，从胸部上方到头部清晰可见；身体向右呈四分之三转身，头部平静地回望镜头。将中棕色头发做成凌乱低发髻，保留少量碎发和细微雀斑，使用自然妆容，穿敞开翻领的哑光米色风衣，背着细暖金属肩带的无标志小单肩包，佩戴一枚低调耳环。用两个大型失焦前景人形框住模特，第三个远处人形和密集人群融入奶油般虚化。以青绿色城市薄雾、普通黄色出租车形状和温暖橙色光斑形成色彩对比。使用柔和阴天光、长焦压缩、大光圈浅景深、锐利眼睛、轻微人群动态模糊和自然皮肤纹理。将以下六条短语逐字、清晰可读地排入不遮挡模特的低调说明文字栏：“成年模特”“4:5竖幅肖像”“米色风衣”“繁忙城市街道”“黄色出租车光斑”“无真实标志”。只允许这些指定短语作为可读文字；不得出现无关标牌或文案、水印或真实标志。不得遗漏、改写或拼错任何指定短语。避免侵入式紧裁、刻板印象、多余肢体和解剖畸变。

English prompt

Use configured identity reference ref-10a-v1 for the clearly adult model. Create an original photorealistic 4:5 vertical street-fashion portrait on a busy generic city street. Place the model slightly left of center, framed from upper chest to head, body turned three-quarters to the right while the head looks back calmly toward the camera. Style medium-brown hair in a messy low bun with a few flyaways, subtle freckles, natural makeup, a matte beige trench coat with open lapels, a small unbranded shoulder bag with a thin warm-metal strap, and one understated earring. Frame the model with two large out-of-focus foreground figures; let a third distant figure and the dense crowd dissolve into creamy blur. Suggest generic yellow taxi shapes and warm orange bokeh against teal-green city haze. Use soft overcast daylight, telephoto compression, wide-aperture shallow depth of field, tack-sharp eyes, gentle crowd motion blur, and natural skin texture. Render these six phrases exactly as crisp, legible editorial typography in a discreet caption band that does not cover the model: “adult model”; “4:5 vertical portrait”; “beige trench coat”; “busy city street”; “yellow taxi bokeh”; “no real logos”. Permit only these specified phrases as readable text; no unrelated signage or copy, watermark, or actual logos. Do not omit, alter, or misspell any required phrase. Avoid invasive cropping, stereotypes, extra limbs, and distorted anatomy.

查看 API 模型

探索更多图片和视频提示词 · 生成图片 · 生成视频 · 在 AI 画布中创作