The Ultimate Nano Banana Prompting Framework

Core Principle: Describe the Scene, Don't List Keywords

Google's official guidance emphasizes that narrative, descriptive paragraphs consistently outperform disconnected keyword lists. The model's strength lies in its deep language understanding, so treat it like communicating with a skilled creative partner.

Essential Prompting Structure

Template Format

[ACTION] + [SUBJECT] + [SPECIFIC DETAILS] + [ENVIRONMENT/SETTING] + [TECHNICAL SPECS] + [PRESERVATION INSTRUCTIONS]

Best Practice Examples

For Image Generation:

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.

For Image Editing:

Using the provided image of [subject], please [add/remove/modify] [element] to/from the scene. Ensure the change is [description of how the change should integrate]. Keep [specific elements to preserve] exactly the same.


The Four Pillars of Effective Nano Banana Prompts

1. Hyper-Specificity is Key

Weak Prompt: "Change the background"
  Strong Prompt: "Change the background to a neon diner at night with pink and blue lighting, vintage chrome fixtures, and subtle steam rising from coffee cups on the counter"

Why This Works: Nano Banana excels when given detailed instructions. The more specific you are, the more control you have over the output.

2. Multi-Turn Editing Strategy

Instead of requesting multiple changes simultaneously, use sequential edits:

Turn 1: "Add a vintage brown leather chesterfield sofa to replace the blue sofa. Keep all pillows, lighting, and room proportions identical."

Turn 2: "Now add a small Persian rug under the coffee table. Match the warm brown tones of the sofa."

Turn 3: "Add subtle warm lighting from a floor lamp in the corner. Keep the natural window light unchanged."

This approach prevents character drift and maintains consistency across edits.

3. Reference Image Integration

When using multiple images, explicitly reference them:

"Place the woman from Image 2 next to the man in Image 1. They sit together, looking at the phone and laughing. Keep cafe lighting and depth of field from Image 1. Match skin tones and reflections to the original scene."

4. Preservation Instructions

Always specify what should remain unchanged:

"Change only the blue sofa to a vintage, brown leather chesterfield sofa. Keep everything else in the image exactly the same, preserving the original style, lighting, and composition."

Specialized Prompting Frameworks

For Photorealistic Results

Think like a photographer and include:

  • Camera specs: "85mm portrait lens," "shallow depth of field," "f/2.8"

  • Lighting: "golden hour light," "soft window light," "dramatic rim lighting"

  • Composition: "close-up portrait," "wide establishing shot," "overhead view"

  • Technical details: "bokeh background," "natural grain," "high contrast"

Example:

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 in his rustic, sun-drenched workshop. Soft, golden hour light streams through a window, highlighting the fine texture of the clay. Captured with an 85mm portrait lens with soft, blurred background bokeh. Vertical portrait orientation.

For Product Photography

Template:

"Replace [original product] with [new product] from Image 2. Match hand pose, reflections, and [material] specular highlights. Keep label readable and preserve text legibility. No stylization."

Example:

"Replace the black can with the orange 'GUERRILLA' can from Image 2. Match hand pose, reflections, and metal specular highlights. Keep label readable and preserve text legibility; no stylization."

For Character Consistency

Lock identity elements:

"Same face, hair, makeup, and earrings across all outputs. Keep [subject] identical while changing [environment/action]. Maintain facial features, expression, and clothing exactly as shown."

For Text Editing

Template:

"Change the text from '[original text]' to '[new text]'. Maintain font weight, curvature, perspective warp, and reflections. Keep brand colors identical. No other changes."


Advanced Prompting Techniques

1. Scene Composition

For complex scene building:

"Create a [mood] scene with [subject] in [environment]. Include [specific elements]. Use [lighting style]. Frame as [shot type]. The atmosphere should feel [emotional tone]. Keep [specific preservation requirements]."

2. Style Transfer Prompts

"Transform this image to [art style] while preserving [specific elements]. Apply [style characteristics] but keep [preservation requirements] unchanged."

3. Environmental Manipulation

"Change the weather to [condition]. Add [atmospheric elements]. Modify lighting to [specification]. Keep all subjects, poses, and clothing identical. Preserve facial features and expressions."


Critical Do's and Don'ts

DO:

  • Be conversational but precise: "Using the provided image of my cat, please add a small, knitted wizard hat on its head"developers.googleblog

  • Specify preservation requirements: "Keep everything else exactly the same"

  • Use reference images: "Use the yellow Porsche from Image 2 as the car"

  • Break complex edits into steps: Edit one element at a timefelloai

  • Include technical photography terms for realism: "bokeh," "golden hour," "shallow depth of field"developers.googleblog

DON'T:

  • Use keyword lists: "cat, hat, wizard, magic"

  • Make multiple changes simultaneously: This causes inconsistency

  • Be vague: "Make it better" or "Change the background"

  • Ignore lighting consistency: Always specify how new elements should integrate

  • Overload with conflicting instructions: Keep prompts focused


Platform-Specific Optimization

Google AI Studio

  • Use the build mode for iterative development

  • Leverage template apps for complex workflows

  • Take advantage of multi-image upload capabilities

Gemini Chat Interface

  • Select "2.5 Flash" model

  • Enable "Create images" tool

  • Use multi-turn conversations for refinement

API Integration

  • Set appropriate aspect ratios: "1:1", "16:9", "9:16"

  • Configure image count (1-4 images)

  • Use structured JSON for complex requests

Pricing Optimization Strategy

At $0.039 per image, optimize your usage:

  1. Start with simple prompts and refine iteratively

  2. Use multi-turn editing instead of regenerating entire images

  3. Batch similar requests to maintain consistency

  4. Save successful prompt patterns for reuse


Troubleshooting Common Issues

If Character Consistency Drifts:

  • Return to the original image and restart the editing sequence

  • Add more specific preservation instructions

  • Use reference phrases like "identical to the original"

If Text Appears Distorted:

  • Add "preserve text legibility; no stylization"

  • Specify "maintain font weight, curvature, and reflections"

  • Include "keep brand colors identical"

If Lighting Looks Unnatural:

  • Specify how new elements should integrate: "match the original lighting"

  • Include directional lighting cues: "soft window light from the left"

  • Add preservation notes: "keep shadows and highlights consistent"


Sample Workflows for Different Use Cases

E-Commerce Product Shots:

  1. "Create a clean white background studio shot of [product]. Even lighting. 3:4 aspect ratio."

  2. "Now place the same product in a modern kitchen setting. Natural lighting from windows."

  3. "Create a lifestyle shot with the product being used by a person. Keep product details identical."

Social Media Content:

  1. "Create a 9:16 Instagram Story version with neon background and space at top for text."

  2. "Now make a 1:1 Instagram post version with clean studio backdrop."

  3. "Create a 16:9 YouTube thumbnail version with cinematic lighting."

Marketing Materials:

  1. "Transform this product photo into a magazine advertisement with urban background."

  2. "Add professional marketing text overlay: '[your message]'. Match brand colors."

  3. "Create three variations with different backgrounds: office, cafe, outdoor."

This framework leverages Nano Banana's unique strengths in contextual understanding, character consistency, and natural language processing to deliver professional-quality results efficiently and cost-effectively.