Build with Nano Banana 2, our best image generation and editing model

Revolutionizing Image Generation and Editing with Nano Banana 2

In a significant update to its AI-powered image generation and editing capabilities, Google DeepMind has unveiled Nano Banana 2, a cutting-edge model that promises to transform the way developers create and manipulate visual content. Built on the Gemini 3.1 Flash Image architecture, Nano Banana 2 boasts improved world knowledge, advanced text rendering, and greater creative control, making it an indispensable tool for a wide range of applications.

Improved World Knowledge: A Leap Forward in Visual Grounding

One of the most significant advancements in Nano Banana 2 is its ability to leverage the Gemini model's broad world knowledge to create enhanced visuals using images via web search. This feature allows developers to generate more detailed depictions inspired by real-life references, opening up new possibilities for creative applications.

To demonstrate the power of this feature, Google DeepMind has built an example app called "Window Seat" that uses Nano Banana 2's knowledge and web image search to create photorealistic window views inspired by world locations and live weather data. This app showcases the potential of Nano Banana 2 to create realistic and engaging visual content that can be used in a variety of contexts, from entertainment to education.

Advanced Text Rendering and Localization: A Boost for Dynamic UI Generators

Another key feature of Nano Banana 2 is its advanced text rendering capabilities, which deliver an upgrade from previous Flash image models. This means that text can now render with the same precision as the artwork, making it an ideal tool for building dynamic UI generators or creative apps.

Nano Banana 2 also supports in-image localization, allowing developers to generate or translate text across multiple languages directly within the image. This feature is particularly useful for international applications, where text needs to be translated and localized for different markets.

To showcase these capabilities, Google DeepMind has built a demo app called "Global Ad Localizer" that translates an advertisement into different languages for international markets. This app highlights the model's advanced in-image localization capabilities, rendering translated text while simultaneously understanding and localizing the visuals.

Greater Creative Control and Consistency: A New Era in Image Generation

Nano Banana 2 delivers speed and visual fidelity, with vibrant lighting, richer textures, and sharper details. Developers now have more control over these high-fidelity outputs with several advanced creative features, including:

Native aspect ratios: Seamlessly generate or edit images to fit project requirements, with native support for all existing aspect ratios.
New 512px resolution: Optimize for efficiency, joining the 1K, 2K, and 4K offerings.
Improved instruction following: The model adheres more strictly to complex, multi-layered developer prompts, ensuring the image requested is generated.
Configurable thinking levels: Adjust thinking levels (Minimal vs. High/Dynamic) to allow the model to reason through complex prompts before rendering, significantly improving output quality and prompt adherence.

To demonstrate these capabilities, Google DeepMind has built a demo app called "Pet Passport" that takes a single photo of a pet and sends them on a global adventure to famous landmarks. This app maintains the pet's appearance across different global destinations, showcasing the model's ability to create consistent and engaging visual content.

Built for Production: What Developers Are Saying

We are already seeing incredible innovation from partners who have integrated the model into their production workflows to scale their visual features. As one developer noted, "Nano Banana 2 has been a game-changer for our team. We can now generate high-quality images at scale, with greater creative control and consistency. It's been a huge time-saver and has opened up new possibilities for our applications."

Build with Nano Banana 2 Today

You can start building with Nano Banana 2 today, using a paid API key with the Gemini API in Google AI Studio, for enterprise deployment on Vertex AI. It is also available for use in Google Antigravity and Firebase. Here are a few places to get started: