What AI model works best with this prompt?

Sora 2 and GPT Image 2 are ideal, as the prompt is designed for subject consistency testing, but you can adapt it for other image generators.

Can I change the subject or the scene?

Yes, the prompt uses variables like `subject`, `location`, and `object`. Simply replace them to test different subjects, settings, or objects.

What kind of output should I expect?

You'll get an image of the specified subject alone in the given location, allowing you to assess how consistently the model renders the subject across generations.

Sam Altman Skateboard Prompt — Test Sora 2 Subject Consisten | sora2hub

GPT Image 2AI 提示词

Sam Altman Skateboard Prompt — Sora 2 Subject Consistency Test

Test AI subject consistency by generating a scene of Sam Altman skateboarding alone in a skate park with this customizable Sora 2 prompt template.

提示词 · EN

argsubject: Sam Altman 在 arglocation: 滑板公园 玩 argobject: 滑板，周围没有其他人。

关于这个提示词

Test AI subject consistency by generating a scene of Sam Altman skateboarding alone in a skate park with this customizable Sora 2 prompt template. 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 subject、location、object 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 template from this page.
Open Sora 2 or your preferred AI image generator (e.g., GPT Image 2).
Modify the variables in curly braces: replace subject, location, and object with your desired terms, or keep the defaults (Sam Altman, skate park, skateboard).
Generate the image and examine the output for subject consistency.
Experiment with different subjects or scenes to thoroughly test model performance.