← Back to galleryDesign
East asian editorial design unbalanced layout poster — After
ChatGPT
DesignChatGPT

East asian editorial design unbalanced layout poster

Creator @MrLarus

Prompt

[Theme]: ____
[Key Visual]: ____
[Chinese Title]: ____
[English Title]: ____
[Three-part Method]: Vertical / Horizontal / 2+1 / Misaligned
[Main Color]: ____
[Auxiliary Information]: ____
[Ratio]: 9:16

Generate a cultural poster in East Asian Editorial Design style, using "unbalanced three-part composition."

Do not split the screen evenly into three blocks; instead, let the three areas respectively carry: key visual, title text, and auxiliary information. The key visual carries the most weight, the title forms a distinct Typography shape, and the smaller area is responsible for time, materials, location, data, maps, or other information truly relevant to the theme.

At least one cross-region relationship should occur among the three areas, such as the main subject crossing boundaries, giant text overlapping the image, branches or structural lines passing through regions, or local textures entering the title area, making the composition have a grid but not be restricted by it.

Use a warm paper background, black text, and one primary accent color; use Modern Songti / Mincho for Chinese, Serif for English, and Grotesk / Monospace for small text. Maintain medium-high information density and retain approximately 20%–30% effective white space.

Add a layer of archival content related to the theme, such as partial enlargement, cross-sections, time records, material information, or diagrams, but do not use random English or template placeholder text.

The same series should have unified paper, fonts, and information systems, with each piece varying the three-part direction, area proportions, theme color, and key visual type.

Avoid: even three-column layout, PPT feel, ordinary full-bleed images with text added, template-swapping, random small text, logos, QR codes, numbering, and copyright markings.

Copy the text, add your own details, and try it in your image model.