neroidea
由 GPT Image 2.5 根据所示提示词生成,点击图片可全屏查看。

随手快照网格拼图

GPT Image 2.5社交配图
提示词来自 @nanohana_hanana

Using REFERENCE_0 as the character base, generate a 3x3 grid collage containing exactly 9 distinct images. The aesthetic should mimic amateur smartphone photography (candid snapshots) rather than polished digital art. Layout & Content: - Create 9 separate panels arranged in a square grid. - Locations: Mix of bedroom (messy background), classroom (desks/chalkboard), and outdoor park (swings/sky). - Outfits: Change the character's clothing to casual streetwear (hoodies, t-shirts, caps) or winter wear (scarves). Do not strictly keep the uniform. - Poses/Actions: Include specific scenarios such as holding a black cat, taking a mirror selfie with a phone covering part of the face, eating a strawberry tart, sitting on a swing, lying down on a desk, and close-ups with fingers partially blocking the lens. Style Constraints: - Apply realistic lighting effects like sun glare, indoor shadows, and slight motion blur to enhance the 'snapshot' feel. - Keep the character's facial features (eyes, hair color/style) consistent with REFERENCE_0 but rendered in a more realistic/anime-hybrid style typical of AI photo generation.

#social post#gpt-image#amateur#snapshot#grid