Google Nano Banana Pro API

Introduction

Nano Banana Pro is Google’s professional image generation and editing model for complex visual briefs, high-fidelity product mockups, accurate text, and controlled creative production. It is designed for tasks where products, people, written content, layout, and visual references must work together inside one polished asset.

On iCreat, you can generate images from text, upload references for guided creation, or edit existing visuals through the same model page. The endpoint supports up to 11 reference images, including six high-fidelity object images and five portrait photos. Choose from 1K, 2K, or 4K output across ten common aspect ratios, then use the model through the online Playground or the google/gemini-3-pro-image API.

Key Features

Professional design and product visuals

Nano Banana Pro is built for polished commercial assets, including product mockups, packaging concepts, campaign graphics, menus, and branded compositions. Its reasoning-driven image workflow helps coordinate products, people, text, materials, and layout inside complex briefs where visual consistency and production-ready detail matter more than rapid, low-cost experimentation.

Controlled multi-reference creation

Upload up to 11 reference images to guide a single result, including as many as six high-fidelity object images and five portrait photos. This structure supports workflows that combine products, packaging, people, clothing, environments, and styles while giving each source a clear role in the final composition.

Accurate text and complex layouts

Nano Banana Pro handles designs where text, imagery, and layout must work together, such as posters, signs, packaging, menus, infographics, and editorial graphics. It is especially useful for briefs that require readable labels, clear hierarchy, multiple visual elements, controlled placement, and a coherent final composition across professional marketing and design assets.

Natural-language image editing

Use written instructions and reference images to revise existing assets without moving to a separate model page. Nano Banana Pro can change backgrounds, lighting, props, styling, or scene details while drawing from uploaded products, portraits, and visual references. The current iCreat endpoint does not expose mask-based editing controls.

Predictable 1K to 4K pricing

Choose 1K, 2K, or 4K output across ten common aspect ratios without selecting a separate quality tier. Both 1K and 2K cost $0.14 per image, while 4K costs $0.24, making per-image budgeting simple for teams producing standardized visual assets across campaigns, channels, delivery sizes, and final formats.

How to Use

1. Choose your workflow

Enter a prompt without uploading an image for text-to-image generation. Upload one or more references when you want to guide a new composition or edit an existing asset.

2. Describe the result

State the subject, setting, layout, text, lighting, and visual style. For edits, clearly separate what should change from the details that must remain consistent.

3. Add reference images

Upload up to 11 references and explain each image’s role, such as product shape, packaging, portrait identity, clothing, environment, or lighting.

4. Select ratio and resolution

Choose a supported aspect ratio and select 1K, 2K, or 4K resolution. Generate in the Playground or call the google/gemini-3-pro-image API.

Pricing

Nano Banana Pro uses fixed per-image pricing based on resolution for text-to-image generation, reference-based creation, and image editing.

Because 1K and 2K cost the same, 2K may offer better value when the workflow can benefit from the larger output.

Model Comparison

Nano Banana Pro vs GPT Image 2

Both models support professional image generation and editing, but they differ most in reference handling, output controls, prompt capacity, and pricing structure.

Parameters

Use Cases

Product mockups and packaging visuals

Combine product, packaging, material, color, lighting, and background references to create polished commercial mockups without relying on a long prompt to describe every visual detail.

Brand campaign assets

Create posters, banners, launch graphics, and campaign variations that bring products, people, brand colors, written copy, and controlled layouts together in one consistent visual direction.

Character and lifestyle scenes

Use portrait, clothing, environment, and style references to place recognizable people in new advertising, editorial, travel, or lifestyle settings while maintaining a coherent visual identity.

Menus, posters, and signage

Generate layouts where readable text, decorative typography, supporting imagery, spacing, and hierarchy must work together, reducing the amount of manual assembly and correction required after generation.

Infographics and visual explanations

Build diagrams, labeled processes, educational graphics, and structured summaries that need clear relationships between text, icons, objects, panels, and supporting visual information for easier reading.

Image editing and asset variations

Modify backgrounds, props, lighting, styling, or scene details while using the original image and additional references to guide new campaign, format, seasonal, or regional variations.

Prompt Tips

Assign every reference a role. State which image controls the product, person, packaging, clothing, background, lighting, or visual style.

Describe the layout. Specify where the subject, headline, labels, and supporting elements should appear instead of requesting only a broad design style.

Quote required text. Put exact wording in quotation marks and keep important text blocks concise enough to review easily.

Separate changes from preserved details. Clearly identify what should change and what must remain visually consistent, especially when editing products, people, or branded assets.

Notes

• The 11-image limit allows up to six object references and five portrait photos; it does not allow 11 references of either type without restriction.
• The current iCreat endpoint supports editing through prompts and reference images but does not expose mask-based or multi-turn conversational controls.
• Google Search grounding is part of the underlying Google model ecosystem but is not exposed through the current iCreat parameters.
• Images generated by the Nano Banana model family include SynthID watermarking.

Related Models

GPT Image 2

Choose it for longer prompts, up to 16 general references, exact output dimensions, and flexible Low, Medium, and High quality pricing.

Nano Banana 2

Choose it when faster iteration and lower-cost image production matter more than Nano Banana Pro’s professional design focus.