WHAT NEW IN X11.02 MON

X 上的 AI 合集档案 · 22 个合集 · 65,263 条

gpt-imageIMAGE

GPT Images 2.5 上下分屏提示词:实景变童年蜡笔画

一条可直接复制的分屏提示词:上半写实风景、下半同构图儿童蜡笔画,中间夹手写纸条。

simeon-sanai
@Naiknelofar7882026.10.01收录原推发布于 2026.09.30
GPT Images 2.5 上下分屏提示词:实景变童年蜡笔画(第 1 张,共 3 张)
GPT Images 2.5 上下分屏提示词:实景变童年蜡笔画(第 2 张,共 3 张)
GPT Images 2.5 上下分屏提示词:实景变童年蜡笔画(第 3 张,共 3 张)
simeon-sanai @Naiknelofar788原推发布于 2026.09.30
在 X 上打开原推

GPT image 2.5 Prompt: Create a visually striking split-scene artwork showing [LANDMARK / CITY / COUNTRY] in two artistic worlds. Top half: a cinematic, photorealistic view of [LANDMARK], captured during a beautiful [morning/golden-hour/evening], with authentic architecture, local people, atmosphere, street details, natural lighting, realistic textures, and strong cinematic depth. Include recognizable cultural details that immediately establish the location. Bottom half: the exact same scene reimagined as a handmade child’s drawing, created with rough colored pencils, wax crayons, imperfect strokes, scribbled textures, uneven outlines, innocent proportions, visible paper grain, playful colors, and charming imperfections. Keep the important landmarks and visual elements recognizable but make them feel lovingly drawn from memory. Between the two worlds, place a small torn piece of cream paper with handwritten text: “[SHORT EMOTIONAL PHRASE]”. The transition should feel like reality turning into a childhood memory. Warm nostalgic atmosphere, tactile paper texture, subtle imperfections, editorial composition, emotionally charming, highly detailed, artistic but believable. No clean digital illustration, no glossy 3D render, no overly polished cartoon. The strongest visual hook is keeping the same composition in both halves so the viewer immediately recognizes that the crayon drawing is the child’s imaginative/memory version of the real place.

x.com/naiknelofar788/status/2105146737359024184采集于 2026.10.01 10:12