What is the best AI model for this prompt?

Sora 2 or other advanced text-to-image models that support mixed realism and stylized 3D rendering work best.

Can I change the number of people or the photo booth style?

Yes, adjust the 'number of people' and 'situation' variables to customize the scene, like 'three friends' or 'vintage photo booth'.

What kind of output can I expect?

Expect a high-fidelity image with realistic lighting and backgrounds, featuring characters rendered in a 3D anime style as if they are close friends posing together.

GPT Image 2AI 提示词

Anime Friends in Photo Booth — Sora 2 Prompt

Create ultra-realistic purikura photos blending 3D anime-style friends with authentic photo booth lighting and backgrounds.

提示词 · EN

生成一张 argnumber of people: 两人 角色在 argsituation: 大头贴机 中拍照的图片，重点在于真实感。背景应呈现真实的实景风格，而角色应采用动漫风格的 3D 渲染，看起来就像亲密的好友一样。

关于这个提示词

Create ultra-realistic purikura photos blending 3D anime-style friends with authentic photo booth lighting and backgrounds. 它更适合作为 GPT Image 2 的概念艺术起点：先保留画面结构、主体关系和风格约束，再替换成你的品牌、人物或场景。

使用时优先替换 number of people、situation 等占位符，并保持镜头、构图和材质描述的顺序不变。这样更容易获得稳定的画面结果，也方便继续做多版本对比。

如何使用这个提示词

Copy the prompt text. 2. Select a compatible AI model (e.g., Sora 2). 3. Replace the number of people and situation variables as desired. 4. Generate the image and review the realistic-anime blend.