google/nano-banana models

nano-banana by Google — AI Model Family

Google's nano-banana family represents a suite of advanced image generation and editing models powered by Gemini, designed for fast, high-quality visual creation and manipulation. Known internally as codenames for Gemini 2.5 and 3.0 preview models like gemini-2.5-flash-image (Nano Banana) and gemini-3-pro-image-preview (Nano Banana Pro), this family solves the need for lightweight, speedy AI tools that deliver precise image outputs without heavy computational demands. Ideal for developers, creators, and everyday users, nano-banana excels in text-to-image generation and image-to-image editing, supporting specialized tasks like sketch conversion, photo enhancement, and style transfers. The family includes 9 models across Edit (Image to Image) and Text to Image categories: Nano Banana Pro | Edit (Image to Image), Nano Banana Pro (Text to Image), Nano Banana | Edit (Image to Image), Nano Banana (Text to Image), Nano Banana Pro - Comic Art (Image to Image), Nano Banana Pro - Photoshoot (Image to Image), Nano Banana Pro - Biometric Photo (Image to Image), Nano Banana Pro - Realism (Image to Image), and Nano Banana Pro - Sketch (Image to Image). These models emphasize speed and fidelity, with Pro variants reaching up to 4K resolution for professional-grade results.

nano-banana Capabilities and Use Cases

The nano-banana family shines in text-to-image and image-to-image editing, with base models optimized for speed and Pro versions delivering enhanced detail and control. Core capabilities include generating images from text prompts, editing existing photos via descriptive instructions (with optional masks), and grounding outputs in real-world data like Google Search results for accuracy.

• Nano Banana (Text to Image) and Nano Banana Pro (Text to Image): Perfect for rapid ideation. Use case: Marketing teams generate product visuals. Example prompt: "A kawaii illustration of the current weather forecast for Paris showing the current temperature in Celsius" – grounded with useGoogleSearchGrounding(true) for real-time data integration.
• Nano Banana | Edit (Image to Image) and Nano Banana Pro | Edit (Image to Image): Transform uploaded images with text edits, ideal for quick fixes like background swaps or restorations. Use case: Restore damaged family photos by prompting "Repair scratches and enhance colors on this old portrait."
• Nano Banana Pro - Sketch (Image to Image): Converts rough sketches to polished images, supporting sketch-to-image workflows. Use case: Designers prototype UI elements from hand-drawn ideas.
• Nano Banana Pro - Realism (Image to Image): Boosts photorealism in edits. Use case: DIY enthusiasts visualize room makeovers, e.g., "Repaint this living room in modern teal with wooden shelves mounted on the left wall."
• Nano Banana Pro - Photoshoot (Image to Image): Crafts professional studio shots. Use case: E-commerce sellers create model poses from casual photos.
• Nano Banana Pro - Biometric Photo (Image to Image): Processes images for ID-compliant formats, handling passport or visa specs.
• Nano Banana Pro - Comic Art (Image to Image): Applies stylized comic effects. Use case: Artists turn photos into dynamic panels for webtoons.

Models support pipelines for chained workflows: Start with Nano Banana (Text to Image) for a base, then refine via Nano Banana Pro - Realism Edit for hyper-detailed outputs. Technical specs include up to 4K resolution on Pro models, flexible aspect ratios (e.g., 1:1), and formats like PNG/JPG with mask support for precise edits. Additional grounding via Google Search or Maps ensures contextually accurate generations, such as location-based visuals.

What Makes nano-banana Stand Out

nano-banana sets itself apart with lightning-fast processing tailored for lightweight deployment, making it Google's go-to for on-device or API-driven image tasks without sacrificing quality. Key differentiators include Google Search grounding, where Nano Banana Pro pulls real-time web references for factual, up-to-date images – like generating a weather illustration backed by current Paris forecasts. Pro models offer high-fidelity 4K outputs and advanced editing precision, maintaining character consistency across generations, which is crucial for series art or photo sequences.

Strengths like speed (optimized Nano Banana for quick previews) and consistency (Pro's style adherence) outperform bulkier alternatives, while features such as flexible visual styles, layout control, and photo restoration enable practical applications from DIY projects to professional shoots. Agentic Vision integration in related Gemini tools enhances detail detection, boosting accuracy by 5-10% through iterative zooming and analysis – a boon for complex edits. This family excels in control and reliability, ideal for developers building apps, content creators needing rapid iterations, marketers crafting visuals, and hobbyists tackling home projects. Its browser integration via Chrome's Gemini sidebar further democratizes access for seamless editing.

Access nano-banana Models via each::labs API

each::labs is the premier platform for harnessing the full nano-banana family through a unified, developer-friendly API at eachlabs.ai. Access all 9 models – from speedy Nano Banana Text to Image to specialized Pro Edit variants like Comic Art and Biometric Photo – with simple integration, no complex setup required. Experiment in the interactive Playground for instant testing of prompts and grounding features, or deploy via our robust SDK for production apps supporting high-volume image pipelines. Sign up to explore the full nano-banana model family on each::labs and unlock Google's lightweight AI for your next project.