提示词模板

SQL Collectible Toy Packaging Grid

Four collectible toy boxes in a 2x2 grid, each a title column, a candy-material character diorama and a blister pack of three props, one per named franchise.

专为 GPT Image 设计插画与艺术第 4 版

从示例开始

点一张示例图，整套设置立即套用，再改成你自己的。

查看案例

改成你的

只展示值得改的部分，提示词会实时更新。

IP 1

First of the four franchises filling the 2x2 grid — one recognizable property, since its era decides that quadrant's palette

IP 2

Second franchise — pick one from a different decade than the first so the four quadrants do not share a retro palette

IP 3

Third franchise — choose one with three obvious props, because the right column turns them into labeled blister-pack miniatures

IP 4

Fourth and last franchise — one with a strong lead character, since the candy-material diorama is staged around that character

你的提示词

SELECT * FROM Collectible_Toy_Packaging WHERE layout_format = '2x2_Quadrant_Grid' AND targets = ARRAY['[IP_1]', '[IP_2]', '[IP_3]', '[IP_4]'] AND quadrant_structure = ARRAY[ (Zone: 'Left_Column', Material: 'Printed_Cardboard', Content: 'Massive_Typography_Title_And_Inferred_Creator_Metadata'), (Zone: 'Center_Stage', Material: 'confection candy', Content: 'infer_main_character_and_diorama(target)'), (Zone: 'Right_Column', Material: 'Transparent_Glossy_Vacuum_Plastic_Blister_Pack', Content: 'infer_three_iconic_props(target)_As_3D_Miniatures_With_Text_Labels') ] AND color_grading = 'Vintage_Retro_Palette_Matching_Inferred_IP_Era' AND camera = 'Product_Photography_Front_Orthographic_View';

点击高亮部分即可直接修改。

适用场景

Use when you want to mock up retro toy packaging for four franchises at once, as fan art or a nostalgia post.

怎样效果最好

Name four franchises from different decades; each box's retro palette is matched to its era, so spread them out.
Pick properties with one lead character and three props; the diorama and blister pack are inferred from the name.
Write each name as it is commonly known; the title and creator credits on the cardboard column are inferred from it.

避坑提醒

Each box adds a title, credits and three prop labels, so twenty text items share one grid; expect garbled small type.
Titles and credits are inferred rather than quoted, so the spelling of a real franchise can drift; check each column.

来源与归属

许可: CC0-1.0
示例案例作者: Gdgtify

继续浏览

更多插画与艺术模板 GPT Image 提示词模板全部模板