A single image organised as a national-museum exhibition panel. Pick an object — Tang dynasty hanfu, Edo-period kimono, Sengoku samurai armor, Ming porcelain, Japanese tea ceremony set, Joseon hanbok, Qing headpiece — name it, pick the annotation language, and the model produces a 5-region poster: header + structure callouts + material samples + palette + assembly diagram. Reads like a museum exhibition panel, not a fashion render.

Heritage object infographic — Tang dynasty Banbi over Ruqun, English annotations on rice-cream paper

What you bring

An object type — hanfu, kimono, samurai armor, Ming porcelain, Japanese tea set, Korean hanbok, Qing headpiece. Determines the entire structural taxonomy and material vocabulary; a hanfu can't read like a kimono, a tea set can't read like a porcelain vase.
An object name — the specific variant: Tang dynasty Banbi over Ruqun, Edo-period Furisode in plum blossom motif, Sengoku-era Tōsei Gusoku, Ming blue-and-white peony scroll vase.
An annotation language — English, Simplified Chinese, Japanese. All callout text, material descriptions, palette names, and assembly steps render in this language consistently.
A paper tone — rice cream, silk white, pale tea, faded bamboo. Sets the background register.

Optionally: an era / period descriptor. Leave it blank and the model infers from the object name.

What you get back

A vertical infographic with five locked regions: title header (with subtitle and intro caption), structure callouts (faceless mannequin / flat-lay / vase profile with thin callout lines naming each component), material & technique swatches, palette + symbolism (5 colour swatches with cultural associations + 3-4 motif samples), and bottom assembly flowchart with 4-6 numbered steps. Reads like a page from a museum catalogue — useful for educators, museum-shop merchandise, fashion-history blogs, and heritage-craft documentation.

Seven traditions, no cross-leak

The prompt forbids tradition-bleed: a hanfu can't render as a kimono, a Ming vase can't read as Greek pottery, a samurai armor can't borrow European-knight cues. Each tradition's component-naming, material vocabulary, and visual register are locked. Real-celebrity faces are explicitly forbidden — the figure is always a faceless mannequin or stylised silhouette.

Cost & timing

8 credits per run. Roughly 70-80 seconds. With your 20 signup credits you get two before topping up.

Try it →

Inspired by @MrLarus on X — the seminal Museum-Style Hanfu Breakdown Infographic showing structure, materials, palette, and assembly in a single annotated panel. The seven-objectType × three-language × four-paper enforcement, locked 5-region layout grid, slot scaffolding, and per-enum FORBIDDEN-cue blocks (no cosplay lighting, no cross-tradition bleed, no celebrity faces) are FluxGen's.