Prompts tagged "Nano Banana Pro" in the PromptCraft library.
Element counts are pinned down exactly (seven pairs of dot-work, two fern leaves), whitespace is written in metric units (at least 8mm between each element), and the finished size is also written in...
`Change only the rendering` cleanly separates 'what to change' from 'what must not change,' and adding...
What makes a composite look fake always comes down to looking 'pasted on' — this line pins down, item by item, that it should compress over muscle bulges, blur slightly at the edges, and pick up the original photo's exposure and...
The key to an infographic is carving out the sections first — quoting each of the three section headers by name in quotation marks keeps the model from blurring text into pure decoration; adding...
Leave 'condensing the information' to the model, but lock down 'which words absolutely must appear' with quotation marks — educational diagrams are most vulnerable to labels going off-script, and this...
Naming `orthographic` along with the three drafting views — plan, elevation, and section — makes the model...
The phrase `deconstructed to show the texture of…` ties an exploded-view diagram together with appetite appeal...
The sketch serves as the layout template, and the model only handles the materials — `Keep the exact layout of…` is...
Clear division of labor — placeholder boxes get 'their content swapped in,' while button positions and the grid are 'not to be touched'; UI mockups are most vulnerable to...
Generate a lively hero storefront image for night market and street-food vendors: warm yellow string lights, sign text accurate character by character in Traditional Chinese, and a blurred crowd of customers queuing in the background — ready to post straight to a Google Business Profile, delivery-platform header, or Facebook page cover.
Generate a lavish key visual for a Ghost Festival (Zhongyuan Pudu) sale: a pyramid-style stack of offerings, warm candlelight, and promotional slogans and dates rendered accurately without garbled text — ready to use for hypermarkets, corner stores, group-buy organizers, and hang-tags.
Kaiju dorsal fins, a blue glowing silhouette, …
Blocking out the sun, casting a giant shadow, …
Subsurface scattering plus three-point lightin…
An exaggerated geometric shape, atomic symbols…
A bulky metal body, big round eyes, and a head…
No masks, no selection tools — just describe in plain language what to change. This prompt is built around the key declaration 'change only this one thing, keep everything else the same,' and the same sentence pattern covers changing backgrounds, removing passersby, or swapping outfits. If a result isn't quite right, just add another line to the same conversation instead of regenerating from scratch — much faster than starting over.
Upload a person photo, a product photo, and a scene photo separately, and clearly specify in the prompt who is the subject, what object is being placed, and what serves as the environment, compositing them into a commercial-looking shot that reads as if it were actually photographed. The key is a declaration for unified lighting and undistorted product labels — great for small e-commerce sellers who can't afford a real model shoot.
The biggest fear when making brand character posts, picture-book pages, or a social series is the character changing by the second image. This prompt separates 'locked' elements from 'variable' ones: the face, fixed features, and style stay locked while only the pose and setting change per image. Generate one image at a time, reusing the same opening instructions, to keep the whole series stable.
When a brand or club needs a mascot IP, what's actually needed isn't a single cute image but a character reference sheet with front, side, and back views laid out together. This prompt locks down the layout requirements for cross-angle consistency, standing pose, and an expression-variation zone, producing a sheet you can use as the reference for every generation that follows.
To generate a street scene that unmistakably reads as Taiwan, the key isn't writing 'Taiwan' — it's naming local elements like covered arcade walkways, scooters, overhead wires, and Traditional Chinese shop signs, then locking the sign text in quotes to prevent typos. Good for localized editorial imagery, presentation scenes, and social media backgrounds.
Tell the model that an image 'will be used on a menu,' and it infers the professional requirements of commercial food photography on its own — an underused piece of official guidance rarely taught in Chinese-language content. The prompt builds in a 45-degree menu-style angle, blank space reserved for copy, and texture-detail declarations; it works for both small eateries and delivery-platform product photos.
Turn a phone selfie into a formal headshot: swap in a solid-color background, formal attire, a front-facing pose, and even lighting, while a preservation clause locks down the facial features so you don't get beautified into someone else. Good for resumes, employee badges, and LinkedIn avatars; most official ID documents require an unretouched real photo, so weigh the risk of rejection honestly before using this for that purpose.
Turn a stretch of story into a single multi-panel manga page: panel choices follow the story's rhythm, there's an emotional arc from panel to panel, characters stay consistent across panels, and dialogue is locked in quotes for accurate Traditional Chinese. It tests consistency, layout, and in-image text all at once — an advanced, all-in-one challenge for Nano Banana Pro.