Miniature crew on a giant dessert
Turn your signature dish into a tiny construction site — a miniature worker crew makes food photos people stop scrolling for.

Original recipe · oimg.ai team · Jun 19, 2026
Original example prompt authored by the oimg.ai team — free to copy and adapt.
- モデル
- gpt-image-2
- アスペクト比
- 3:4
- カテゴリ
- フード・ドリンク
変数を入れ替える
画像ではなく構造を拝借しましょう。ハイライトされた各{{TOKEN}}を自分の被写体に置き換えて、生成します。
A macro photography scene of {{DISH}} on {{PLATE_OR_SURFACE}}, treated as a giant construction site by a crew of tiny miniature figurines: {{WORKER_ACTION_1}}, {{WORKER_ACTION_2}}, and a foreman standing on {{FOOD_DETAIL}} pointing upward at the work. Set on {{SETTING}}. Lit by bright natural window light from the left with a gentle fill, giving clean appetizing highlights. Captured with a 100mm macro lens at f/4, shallow depth of field focused on the figures, tilt-shift miniature feel. Playful commercial food photography for a social media post, no watermark, no extra text.
効果的な理由
- The scale contrast between named tiny workers and one oversized dish is a built-in curiosity gap — the eye snaps to the figures first, then re-reads the food, and that double-take is exactly what stops a scroll.
- Giving every figurine a specific job (wheelbarrow, rappelling, pointing foreman) forces the model to compose a coherent little narrative instead of sprinkling random tiny people that float or melt into the frosting.
- The 100mm macro at f/4 with focus pulled to the figures does the miniature illusion for real — shallow depth of field is how actual tilt-shift photography sells scale, so the camera line is load-bearing, not decoration.
- Bright window light keeps the dish looking appetizing; moody dramatic lighting would cue 'art project' instead of 'food you want to eat', which is what makes this usable for cafes and bakers, not just a gag.
自分だけのバージョンを作る
- 1プロンプトをコピー
- 2自分の被写体やディテールに入れ替える
- 3生成する
注意点
- A vague 'tiny people on food' with no assigned jobs tends to produce floating or half-melted figures — the specific worker actions are what hold the scene together.
- Asking for sharp front-to-back focus contradicts the tilt-shift look; the shallow focus on the figures is what sells the scale, so leave it in.
関連レシピ
すべてのプロンプトを表示する
Nano Banana ProHyper realistic edible version of uploaded product
X の @AvelyrahnAI が共有した実例プロンプト
GPT Image 2The bubble tea starts to slowly spin while the ingredients s
X の @Strength04_X が共有した実例プロンプト
GPT Image 23 on @Hailuo_AI
X の @Strength04_X が共有した実例プロンプト
GPT Image 2Ultra-stylized fashion editorial poster
X の @Taaruk_ が共有した実例プロンプト
GPT Image 2Prompt ⤵
X の @miratechtool が共有した実例プロンプト
GPT Image 2Transform the uploaded brand logo image into a premium food-
X の @AIwithkhan が共有した実例プロンプト
GPT Image 2Image prompt by @FutureVibesAi
X の @FutureVibesAi が共有した実例プロンプト
Nano Banana ProA hyper-realistic full-body portrait of a cute young woman c
X の @ariaxawan が共有した実例プロンプト