Nano banana prompts are structured text instructions that direct AI image models like Gemini to generate or transform images with specific subjects, styles, lighting, and camera settings. This guide gives you a working library of copy-paste prompts for portraits, headshots, style transfers, and 3D collectible figures, plus the syntax rules that make them reliable.
What Makes a Nano Banana Prompt Work
Nano banana models respond to concrete, layered descriptions. A prompt that says "a woman in a garden" produces a generic result. A prompt that specifies subject, wardrobe, lighting direction, lens focal length, and aspect ratio produces an image you can use. The model does not guess intent; it renders what you describe.
Every prompt in this library follows the same skeleton:
- Subject: who or what, with age, expression, and pose.
- Wardrobe/Props: specific garments, textures, or objects.
- Environment: location, time of day, weather.
- Lighting: source, direction, quality (soft, hard, rim, backlight).
- Camera: lens (85mm, 35mm), aperture (f/1.8), angle (eye-level, low).
- Style: film stock, art movement, render engine.
- Aspect ratio: 16:9, 4:5, 1:1, 9:16.
This structure is not arbitrary. It mirrors how photographers and art directors brief a shoot. For a deeper breakdown, see the prompt structure and syntax guide.
Nano Banana Prompt Library: Core Examples
These prompts are ready to copy into your model. Replace bracketed variables with your own details.
Portrait Transformation
Use this to turn a casual photo into a studio portrait while preserving identity.
"Transform the subject into a professional studio portrait. Soft key light from 45 degrees left, white bounce fill from right. 85mm lens, f/2.0, eye-level. Subject wears a charcoal turtleneck, neutral expression, hair pulled back. Background: seamless dark gray. Color grade: neutral with slight warmth. Aspect ratio 4:5."
For more portrait variations, the nano banana portrait prompts collection covers mood, age, and expression control.
Professional Headshot
"Corporate headshot of a person in their 30s, wearing a navy blazer over a white shirt. Lighting: large softbox above and slightly forward, subtle rim light on hair. 105mm lens, f/4, chest-up framing. Background: blurred office with warm bokeh. Skin texture natural, no heavy retouching. Aspect ratio 1:1."
This prompt pairs well with the AI headshot prompts guide for LinkedIn and resume use.
Style Transfer: Oil Painting
"Repaint the subject as a 19th-century oil portrait. Visible brushstrokes, impasto texture on fabric. Palette: burnt sienna, ochre, deep umber. Lighting: single candle from left, warm falloff. Canvas texture visible. Aspect ratio 3:4."
See oil painting portrait prompts for more art-historical styles.
3D Collectible Figure
"Render the subject as a 3D collectible vinyl figure. Smooth matte plastic finish, exaggerated head-to-body ratio 1:3. Displayed on a circular base with a nameplate. Studio lighting: two softboxes, white background. Isometric angle, 50mm lens. Aspect ratio 1:1."
The AI 3D action figure prompts article covers articulation, packaging, and scale.
Cinematic Film Still
"Cinematic still of a detective in a rain-soaked alley at night. Neon signs reflect in puddles. Anamorphic lens flare, 2.39:1 aspect ratio. Lighting: practical neon from right, cool moonlight from above. Film grain: Kodak Vision3 500T. 35mm lens, f/1.4."
For more film looks, see cinematic film stills prompts.
Nano Banana Prompt Examples for Common Tasks
Product Photography
"Place the product on a white acrylic surface. Lighting: large scrim from top, two strip boxes on sides for edge definition. 100mm macro lens, f/8, focus stacked. Reflection subtle. Aspect ratio 4:5."
More setups in AI product photography prompts.
Fashion Editorial
"Full-body fashion editorial. Model wears an oversized wool coat, standing against a concrete wall. Lighting: overcast daylight, soft shadows. 50mm lens, f/2.8, low angle. Color grade: desaturated with cool shadows. Aspect ratio 4:5."
See fashion editorial prompts for pose and styling direction.
Fantasy Character Design
"Fantasy character design: elven ranger, leather armor with leaf motifs, quiver on back. Lighting: dappled forest light from above. 85mm lens, f/2.0, three-quarter view. Background: ancient trees, mist. Aspect ratio 2:3."
The fantasy sci-fi character design guide covers armor, weapons, and world-building.
Architectural Interior
"Modern living room interior. Floor-to-ceiling windows, morning light casting long shadows. 24mm lens, f/8, eye-level. Materials: oak floor, linen sofa, marble coffee table. Color palette: warm neutrals. Aspect ratio 16:9."
More in architectural interior design prompts.
How to Build Your Own Nano Banana Prompt Library
A personal library saves time and keeps results consistent. Organize by use case, not by model version.
1. Start with a Base Template
Write one prompt per category: portrait, product, landscape, character. Save them as text snippets. When you need a variation, change only the variables: lighting, lens, aspect ratio.
2. Use Negative Prompts Sparingly
Nano banana models handle negatives differently than Stable Diffusion. Instead of a long negative list, state what you want. For example, "clean background, no clutter" works better than "no trees, no cars, no people." The negative prompts guide explains when negatives help.
3. Lock Aspect Ratio Early
Aspect ratio changes composition. Decide 16:9 for landscapes, 4:5 for portraits, 1:1 for products. Put it at the end of the prompt so the model applies it last.
4. Test One Variable at a Time
If a result misses, change only the lighting or only the lens. Changing three things at once makes it impossible to know what worked.
5. Document What Works
Keep a simple text file with prompt + result notes. Over time, you build a personal nano banana prompt library that reflects your style.
Nano Banana Prompt Guide: Advanced Techniques
Lighting Control
Lighting is the highest-leverage variable. Specify direction (left, right, above, behind), quality (soft, hard, diffused), and color (warm, cool, neutral). "Rembrandt lighting" or "butterfly lighting" are recognized terms. For a full breakdown, see mastering AI portrait lighting prompts.
Camera and Lens Language
Mentioning a lens changes perspective. A 24mm lens exaggerates depth; an 85mm lens compresses features. Aperture controls background blur: f/1.4 gives heavy bokeh, f/8 keeps everything sharp. The bokeh and depth of field guide covers this in detail.
Style Fusion
Combine two styles for unique results: "watercolor and ink wash" or "cyberpunk and art deco." Keep the fusion to two styles; three or more becomes muddy. The style fusion techniques article shows how to balance them.
Color Grading
Specify a film stock or color palette. "Kodak Portra 400" gives warm skin tones. "Teal and orange" gives a blockbuster look. "Desaturated with cool shadows" gives a moody edit. The color theory for AI images guide explains palettes.
Common Mistakes and Fixes
Mistake: Too vague. "A beautiful landscape" gives a generic result. Fix: "Mountain lake at sunrise, mist on water, 24mm lens, f/11, golden hour."
Mistake: Conflicting lighting. "Soft light and harsh shadows" confuses the model. Fix: choose one quality.
Mistake: Ignoring aspect ratio. The model defaults to square or 16:9. Fix: always state it.
Mistake: Overloading the prompt. More than 60-80 words often dilutes focus. Fix: prioritize the five most important details.
Mistake: No subject anchor. For transformations, describe the subject's key features to preserve identity. The portrait age features control article covers this.
Where to Go Next
If you are new to nano banana models, start with portraits and headshots. They teach lighting and lens control quickly. Then move to style transfers and 3D figures. The nano banana pro use cases article shows commercial applications.
For a broader view of AI image trends, see AI image generation trends 2026.
Try NanoPrompts
NanoPrompts is a library of ready-to-use AI image-generation prompts and templates. It provides copy-paste prompts for photo transformations such as portraits, headshots, style transfers, and 3D collectible figures. If you want to skip the trial and error, browse the library and paste a prompt into your model. Visit NanoPrompts to get started.
