Labubu in Motion

Camera: High-end fashion magazine spread. Left 60% is the realistic urban street-snap hero image; right 40% is a utility sidebar. Separate them with brush strokes, torn paper, or f…

Model recorded for this example: gpt-image-2.5

Complete prompt

Camera: High-end fashion magazine spread. Left 60% is the realistic urban street-snap hero image; right 40% is a utility sidebar. Separate them with brush strokes, torn paper, or fluid geometric dividers. Shallow depth of field, crisp subjects. Accurate readable headlines: “MONSTER STYLE”, “CITY TWINS”, “【CHARACTER】 x 【USER NAME】”. Scene: Realistic urban street with a blurred background. Sunlight or streetlight must hit both subjects from the same direction so they share one physical space. Subject: Strictly preserve the real user’s facial features, hair color/style, body proportions, identity, and every clothing detail from the reference, such as khaki jacket, plaid lining, lace tunic, split-pattern tie, and newsboy cap. Natural confident street-snap expression. Select a Pop Mart The Monsters character by outfit vibe: Labubu for playful casual, Zimomo for cool edgy, Tycoco for avant-garde structural looks. Dress it in a custom miniature version of the user’s outfit, tailored to its ears, neck, and body. Render hyper-real plush fur or matte vinyl with realistic texture. Action: Human and character must actively interact, never simply stand side by side: high-five, character on shoulder, user fixing its tie, holding hands while walking, or looking at a phone together. Character scale is knee-high when walking or shoulder-sized when carried, obeying believable toy physics. Lighting / Color: Coherent lighting and contact shadows. Premium commercial fashion photography, contrasting realistic textiles with plush or vinyl materials. Sidebar includes mood tags “Situation: Coffee Run / Mood: Cheeky / Vibe: Retro Workwear”, a five-swatch outfit palette with HEX/Pantone codes, bold “OOTD STYLE”, and five ghost-mannequin item cutouts with Chinese names and ¥ prices. Negative Prompt: do not alter the user’s identity, face, hair, body proportions, or reference outfit; no detached interaction, wrong scale, floating, inconsistent lighting, plastic human skin, low resolution, malformed hands, extra fingers, garbled text, misspellings, cluttered sidebar, watermark, or unrequested logos.

Results vary by model and settings. Keep the subject and composition, then adjust lighting, materials or background for your own project.