GPT Image Prompt
Create a premium lock screen wallpaper for iPad Pro in 4:3 ratio, target production canvas 2048x2732 px and required final delivery size 2048x2732 px, inspired by the uploaded image. Transform the source image into a clean, usable iPadOS wallpaper while preserving its core visual DNA: a tiny doll-like child character in a long white textured dress, oversized woven straw sunhat, round rosy face, bright blue eyes, calm smile, turquoise stucco-like wall and floor, strong soft-edged shadow cast to the right, sunny minimalist mood, clay-toy / soft 3D illustration texture, and a cheerful aqua, cream, coral, and sky-blue palette.
Design the wallpaper for an iPad Pro display family, suitable for Liquid Retina XDR or newer Tandem OLED behavior, with large smooth color fields, refined highlights, soft premium contrast, clean luminous aqua tones, and no overcooked brightness. Keep the iPadOS status, widget, clock, and dock areas readable in both portrait and landscape crops. Place the main visual focus near the center with generous surrounding space, keeping the character small but clear and not touching any edge. Preserve a wide calm turquoise negative-space field above and around the hat, keep the bottom area low-detail for dock readability, and avoid dense texture behind widgets or icons.
Use a soft cute icon-friendly 3D character wallpaper style with pastel coastal color grading, smooth sunlit shadows, subtle wall texture, refined tablet-scale detail, and a balanced crop-safe composition. The final image should feel premium, clean, high-resolution, visually memorable, gentle, playful, and actually usable as an iPad wallpaper. No text, no logo, no watermark, no fake UI, no app icons, no phone mockup. If the generator exports a different pixel size, resize or upscale to the required final delivery size while preserving the target aspect ratio and safe areas.
Negative Prompt:
low quality, blurry, noisy, pixelated, watermark, logo, text, signature, fake UI, app icons, phone mockup, border, frame, cluttered composition, busy top area, unreadable clock area, messy icon background, important details near screen edges, cropped subject, harsh neon bloom, large pure-white areas, noisy gray-black gradients, oversaturated colors, distorted subject, poor composition, distorted face, broken anatomy, extra limbs, deformed hands, unnatural eyes, creepy doll expression, cropped hat, tiny unreadable character, character touching edges, muddy turquoise, dirty wall texture, banding, chaotic pattern, high-frequency detail everywhere, overexposed highlights, harsh shadow edges, busy dock area
How to use: Copy and paste this prompt into ChatGPT, Gemini, Nano Banana, or your favorite AI image tool. If it mentions LoRA, reference image, uploaded photo, or identity reference, upload your own photo first. No reference photo? Replace the character with a clear age, gender, and nationality, such as “25-year-old Japanese woman” or “30-year-old American man”.