I turned a real city into GPT Image 2 prompt

1 / 2

Nexora

Author @frametheory058

View on

90Likes6.7KViewsMay 19, 2026

Summary

I turned a real city into a LEGO-style miniature universe Made with GPT Image 2 on @openart_ai

Prompt

I turned a real city into a LEGO-style miniature universe Made with GPT Image 2 on @openart_ai Prompt: Use the uploaded reference image as the primary character reference. Create a premium vertical 4:5 editorial illustration of the same character from the reference image, sitting at a cozy craft table and building a LEGO-style miniature diorama of [CITY NAME]. Only the character remains organic and natural. Everything else must be built entirely from visible LEGO-style bricks: landmarks, streets, bridges, rivers, lakes, trees, vehicles, trains, people, cafés, shops, food stalls, parks, signs, boats, and tiny city details. Preserve the character’s exact identity, facial structure, face shape, eye shape, hairstyle, skin texture, age impression, expression, mood, and overall styling from the reference image. The character must NOT look at the camera. Instead, the character is leaning forward naturally while looking toward the LEGO city and carefully placing or adjusting a tiny brick-built landmark with a calm, proud, imaginative expression. Create a charming handmade LEGO city that feels crafted by a talented child: visible studs, interlocking bricks, playful proportions, slightly imperfect geometry, simplified cute architecture, layered depth, colorful but balanced details, tactile toy-like realism. Include iconic landmarks and symbols of [CITY NAME]: [LANDMARK LIST] Add rich local city details appropriate to [CITY NAME]: tiny LEGO people walking, buses, subway or trains, taxis, bicycles, cafés, local shops, food stalls, street lamps, traffic lights, crosswalks, flower beds, trees, parks, rivers or waterfronts, bridges, boats if relevant, neighborhood scenes, billboards, local street signs, cultural elements, and recognizable city atmosphere. Composition: Vertical 4:5 composition. Character positioned in the upper half of the frame. Large LEGO city diorama filling the lower half. Layered cinematic depth with foreground streets/water/details, middle-ground landmarks, and background skyline. Rich and highly detailed but clean, organized, and readable — never chaotic. Lighting: Bright neutral daylight, soft diffused morning sunlight, clean white balance, crisp whites, gentle soft shadows, airy Scandinavian children’s magazine mood, fresh atmosphere, pastel-friendly LEGO colors, realistic ambient bounce light. Style: Premium children’s editorial illustration, stylized realistic 3D character, LEGO-style miniature world, cozy family-home craft room, whimsical storytelling, ultra-detailed toy photography aesthetic, cinematic depth of field, soft premium textures, polished magazine-quality composition. Text: Bottom-left very tiny watermark: “[Paige.Prompts]” Bottom-right handwritten white crayon text: “Let’s play in [CITY NAME]!” Negative prompt: no Play-Doh, no clay, no papercraft, no realistic architecture, no professional scale model, no changed identity, no different hairstyle, no wrong age impression, no looking at camera, no yellow tint, no orange cast, no sepia, no dark lighting, no messy chaos, no distorted anatomy, no extra fingers, no fake text, no random letters, no logos, no watermark except [Paige.Prompts] [1/1] @openart_ai http://openart.ai/home?utm_source=tolt&utm_medium=affiliate&utm_campaign=affiliate-tolt--acq-web&ref=openartaihomeutmsourcetoltutmmediumaffiliateutmcampaignaf

Model: GPT Image
Categories: Creative Visual
License: CC BY 4.0

Use this Prompt