信息图与卡片

自然观察风食品剖面

在深色展示板上并置完整食物与切面,只标注可见结构,不编造科学数据

更新日期:

完整提示词

英文提示词包含建议的示例需求。生成前请修改主体和细节。原始示例输入和生成参数尚未完全核实。

推荐模型: GPT Image 2.5 Flare

Compose a photographic specimen plate showing one familiar food in two matched views: its intact exterior and a clean longitudinal half-section at the same scale. If reference images are provided, preserve their variety, ripeness, skin pattern, and visible internal arrangement; otherwise use the recognizable everyday structure of the named food. Position the views with enough space to compare their outlines, resting on a matte charcoal surface rather than floating without support. A narrow side light should separate glossy flesh, rough peel, cavities, seeds, and any fibrous areas without making the food look metallic. Keep the cut face sharp and let the background fall into darkness. Add a restrained title and up to four short annotations linked by thin leader lines to visible features. Use ordinary descriptive names only; exclude nutrient values, chemical percentages, health claims, and invented anatomical terminology. Leave out any label whose location cannot be identified confidently. Include gentle surface moisture where appropriate, but no decorative splashes or additional ingredients. Present a quiet naturalist study that helps a viewer compare the outside with the inside, with typography comfortably separated from the specimen. User request: Study a ripe avocado. Show the intact dark pebbled exterior beside a longitudinal half with pale green flesh and a brown seed, both at matching scale. Use the title "AVOCADO STUDY" and only these labels where visible: "Peel", "Flesh", "Seed". No numerical data.

提示词技巧

  • 选择内部结构容易辨认的食物;特殊品种可上传切面参考图,并逐条核对标注是否对应可见部位

来源与署名

由 OmniArt 改编和重写,以下列出来源署名、修改说明及使用条件

  • Naturalist Food Specimen Cross-Section · @GeekCatX

    EvoLinkAI/awesome-gpt-image-2-API-and-Prompts

    The upstream repository is licensed under CC0 1.0, which does not establish rights to third-party community prompts or media. This template is independently authored; no source text or media is copied.

    Independently authored from a general workflow concept. Compares intact and cut views at matching scale; annotations describe visible structures without invented scientific data. No source prompt text or media copied. GPT Image 2.5 Flare is a recommendation, not a verified example model.

来源与署名

相关提示词配方