OpenZhenZhen AI プロンプトギャラリー · gpt-image-2.5 · image

週末の九宮格ライフフォトコレクション

Weekend Nine-Grid Life Photo Collection

「週末の九宮格ライフフォトコレクション」は gpt-image-2.5 を使用した作品です。OpenZhenZhen で元のプロンプトと生成例を確認し、創作の参考にできます。

週末の九宮格ライフフォトコレクション — AI image creation
週末の九宮格ライフフォトコレクション

プロンプト

中国語プロンプト

生成一张1:1方形、由九张不同生活照片组成的3×3周末照片合集,像准备开启个人社交账号时整理的一组自然随手拍;只生成单张完整拼贴,不生成平台界面、账号页、边框按钮或商标。九格各自构图清楚、分隔整齐、无文字标签,并按顺序表现:1镜前自拍;2咖啡馆桌边;3与朋友共进晚餐;4轻微运动模糊的室内聚会瞬间;5城市街道散步;6笔记本电脑与咖啡的工作片刻;7与一只{pet_type}相处;8温暖的日落;9不经意的开怀笑容。每格都出现同一位明确21岁以上的成年人,面部身份和外貌绑定ref-10a-v1,穿着与{weekend_mood}相符的日常服装;聚餐朋友也均为成年人,面部如可见同样使用ref-10a-v1,避免出现未成年人或可识别的额外身份。用不同景别、自然环境光、轻微手机摄影颗粒和真实随拍构图体现{weekend_mood};除第4格有受控的轻微动作模糊外,其余主体保持清楚。宠物、食物与场景不暗示未经提供的真实事件。九格内容必须齐全且互不重复,不得漏格、增格、拼错人物身份或加入水印、文字、应用UI、品牌标志、未成年人或危险饮酒画面。

英語プロンプト

Create one square 1:1 image as a neat 3×3 collage of nine distinct weekend snapshots, like a natural personal photo dump prepared for a new social account. Produce a single complete contact sheet only; do not render a platform interface, profile page, buttons, borders with UI, or social-media logos. Keep every tile separately composed and free of text labels. In order, show: 1 mirror selfie; 2 café table moment; 3 dinner with friends; 4 a slightly motion-blurred indoor party snapshot; 5 a walk along a city street; 6 a laptop-and-coffee work moment; 7 time with a {pet_type}; 8 a warm sunset; 9 an unposed burst of laughter. The same clearly adult person, age 21 or older, appears in every tile, with face identity and appearance bound to ref-10a-v1 and casual clothing suited to {weekend_mood}. Dinner companions are adults too; if their faces are visible, bind them to ref-10a-v1 as well, and include no minors or extra identifiable identities. Use varied framing, natural available light, subtle phone-camera grain, and candid compositions to convey {weekend_mood}. Keep the subject clear in every tile except for controlled slight motion blur in tile 4. Do not imply that the scenes document verified real events. Include all nine distinct tiles with none missing or duplicated; do not alter the recurring subject's identity or add watermarks, text, app UI, brand marks, minors, or unsafe drinking scenes.

API モデルを見る

画像・動画プロンプトをもっと見る · 画像を生成 · 動画を生成 · AI キャンバスで作成