Which AI model works best for this prompt?

Sora 2 produces the most detailed results, but Midjourney and DALL·E 3 also handle grid layouts well.

Can I change the grid size or expressions shown?

Yes — edit the grid variable (e.g., 5×5 or 10×10) and specify any additional expression keywords you need.

Will each cell include readable AU codes?

At higher grid resolutions, AI models generally render legible AU labels, though very large grids may require upscaling for clarity.

FACS Face Expression Grid Chart — 49 AU Code Prompt | sora2hub

GPT Image 2AI 提示词

FACS Face Expression Chart — 49 Expressions Grid Sora 2 Prompt

Generate a comprehensive 49-expression grid chart with Hollywood FACS Action Unit codes for visual reference, training, or creative projects.

提示词 · EN

生成一张包含 arggrid: 7x7 共 49 种图案的网格 的单张图像，展示各种面部表情及其对应的好莱坞 AU 代码（面部动作编码系统）。

关于这个提示词

Generate a comprehensive 49-expression grid chart with Hollywood FACS Action Unit codes for visual reference, training, or creative projects. 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 grid 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 full prompt text from the page.
Open your preferred AI image generator (Sora 2, Midjourney, DALL·E, etc.).
Paste the prompt and adjust the grid variable if you want a different layout size.
Generate the image and review the expression details.
Re-run or tweak keywords until the grid displays clearly.