Nano Banana is Google's image generation and editing model, available inside Gemini and Google AI Studio, and the fastest way to get good output from it is to stop describing pictures and start filling in templates. The six below cover the jobs people actually need images for: photorealistic scenes, stickers, logos and text, product mockups, minimalist backgrounds, and comic panels. Each one comes with a fill-in-the-blank template and a worked example. We test this kind of thing for 300,000+ readers at AI Central, and the pattern holds across every image model: structure beats adjectives. More prompt packs like this one sit in the AI Central Library.
How Nano Banana prompting works
You can either describe what you want in one go, or chat with the model to create, edit, and combine images across several turns. Both work. The difference between a vague result and a usable one is almost always specificity: shot type, lighting, materials, aspect ratio, and what the background should be doing.
Every template below uses square brackets for the parts you replace. Fill in all of them. A half-filled template produces a half-considered image.
1. Photorealistic scenes
For realistic images, think like a photographer. Naming camera angles, lens types, lighting, and fine detail pushes the model toward a photographic result rather than an illustrated one.
Template: A photorealistic [shot type] of [subject], [action or expression], set in [environment]. The scene is illuminated by [lighting description], creating a [mood] atmosphere. Captured with a [camera/lens details], emphasizing [key textures and details]. The image should be in a [aspect ratio] format.
Example prompt: A photorealistic close-up portrait of an elderly Japanese ceramicist with deep, sun-etched wrinkles and a warm, knowing smile. He is carefully inspecting a freshly glazed tea bowl. The setting is his rustic, sun-drenched workshop. The scene is illuminated by soft, golden hour light streaming through a window, highlighting the fine texture of the clay. Captured with an 85mm portrait lens, resulting in a soft, blurred background (bokeh). The overall mood is serene and masterful. Vertical portrait orientation.
2. Stylized illustrations and stickers
For stickers, icons, or asset packs, be explicit about the style and remember to ask for a white background if you need one to cut out.
Template: A [style] sticker of a [subject], featuring [key characteristics] and a [color palette]. The design should have [line style] and [shading style]. The background must be white.
Example prompt: A kawaii-style sticker of a happy panda wearing a tiny bamboo hat. It's munching on a green bamboo leaf. The design features bold, clean outlines, simple cel-shading, and a vibrant color palette. The background must be white.
3. Accurate text in images
Text rendering is one of the model's genuine strengths, which makes it viable for logos and simple brand assets. Be clear about the exact text, the font style described in words, and the overall design direction.
Template: Create a [image type] for [brand/concept] with the text "[text to render]" in a [font style]. The design should be [style description], with a [color scheme].
Example prompt: Create a modern, minimalist logo for a coffee shop called "T-ime to go". The text should be in a clean, bold, Anton font. The design should feature a simple, stylized icon of a coffee bean seamlessly integrated with the text. The color scheme is yellow and black.
4. Product mockups
This is the one that saves real money. Clean, studio-lit product shots for e-commerce, ads, or brand decks, without a studio booking.
Template: A high-resolution, studio-lit product photograph of a [product description] on a [background surface/description]. The lighting is a [lighting setup, e.g., three-point softbox setup] to [lighting purpose]. The camera angle is a [angle type] to showcase [specific feature]. Ultra-realistic, with sharp focus on [key detail]. [Aspect ratio].
Example prompt: A high-resolution, studio-lit product photograph of a sleek stainless steel water bottle in a brushed silver finish, presented on a smooth, dark wooden surface. The lighting is a three-point softbox setup designed to create soft, diffused highlights and eliminate harsh shadows. The camera angle is a slightly elevated 45-degree shot to showcase its ergonomic design. Ultra-realistic, with sharp focus on the subtle condensation droplets on its surface. Square image.
5. Minimalist and negative space design
Use this when the image is not the point. It produces backgrounds for websites, presentations, and marketing materials where text will sit on top, and the empty space is the deliverable.
Template: A minimalist composition featuring a single [subject] positioned in the [bottom-right/top-left/etc.] of the frame. The background is a vast, empty [color] canvas, creating significant negative space. Soft, subtle lighting. [Aspect ratio].
Example prompt: A minimalist composition featuring a single, tiny, unopened pink cherry blossom bud positioned delicately in the mid-right of the frame. The background is a vast, empty soft pastel green canvas, creating significant negative space for text. Soft, diffused lighting from the top left. Square image.
6. Sequential art: comic panels and storyboards
Build visual narratives one panel at a time. Useful for storyboards, comic strips, and any sequential art where the scene description carries the whole load.
Template: A single comic book panel in a [art style] style. In the foreground, [character description and action]. In the background, [setting details]. The panel has a [dialogue/caption box] with the text "[Text]". The lighting creates a [mood] mood. [Aspect ratio].
Example prompt: A single comic book panel in a gritty, high-contrast noir style. In the foreground, two silhouetted figures face each other, one handing over a small glinting object. In the background, a rain-slicked alley with overturned trash cans and a broken window. The panel has a caption box in the top left with the text "Some nights, the shadows had teeth". The lighting creates a tense, moody atmosphere. Square image.
Getting better results from any of these
A few habits that raise the hit rate regardless of which template you start from:
Iterate in conversation rather than rewriting from scratch. Generate, then ask for one change at a time.
Describe fonts and styles in words, not by name alone. "Clean, bold, geometric sans-serif" travels further than a font name the model may not know.
Always state the aspect ratio. Left unspecified, you get whatever the default is, and then you crop.
Say what the background must do. White for cutouts, empty for text overlay, blurred for product focus.
The underlying skill here is prompt structure, not image jargon, and it transfers everywhere. The 26 principles of prompt engineering covers the patterns that make any model easier to steer. If you are generating minimalist backgrounds for slides specifically, pair this with our AI prompts for better presentations.
Frequently asked questions
What is Nano Banana?
Nano Banana is Google's image generation and editing model. You can use it to create images from a text description, edit existing ones, and combine multiple images, working either in a single prompt or through a back-and-forth conversation.
Where can I use Nano Banana?
It is available through Gemini and Google AI Studio. Google AI Studio gives you more control over settings, while Gemini is the faster route for casual generation and editing.
How do I write a good Nano Banana prompt?
Start from a template rather than a blank line. Name the shot type, the lighting, the materials or textures, the background behaviour, and the aspect ratio. Specificity in those five areas accounts for most of the difference between a weak result and a usable one.
Can Nano Banana render text and logos accurately?
Text rendering is one of its stronger capabilities, which makes it practical for logos and simple brand assets. Quote the exact text you want in the prompt and describe the font style in words rather than relying on a font name.
Do these prompt templates work with other image models?
Largely, yes. The structure - subject, action, environment, lighting, camera, aspect ratio - is model-agnostic, so the same templates carry over to other image tools with minor wording changes. More tested prompts are in the AI Central Library.
