Google Unveils Nano Banana 2 Neural Network

The company also unveiled a major update to the Flow service.

Google has launched Nano Banana 2, a new model for visual generation operating at Gemini Flash speed.

Introducing Nano Banana 2: Our best image generation and editing model yet. 🍌

Pro-level quality, at Flash speed. Rolling out today across @GeminiApp, Search, and our developer and creativity tools. pic.twitter.com/6oNWYhVSqp

— Google (@Google) February 26, 2026

The product leverages the Gemini knowledge base and real-time web search data. This approach enables the neural network to more accurately render complex objects, generate infographics, and transform text notes into visual diagrams.

Nano Banana 2 can also generate clear and readable text within images, from marketing layouts to greeting cards. Its built-in localization feature allows text adaptation for any audience.

The model generates detailed photorealistic images without sacrificing speed. Key differences of Nano Banana 2 from its predecessor include:

• Object consistency: The neural network maintains the appearance of up to five characters and the accuracy of 14 objects within a single project;
• Execution precision: The system better understands complex text prompts and considers minor nuances, strictly following user requests;
• Format flexibility: Full control over aspect ratios and resolutions from 512px to 4K, from vertical posts to widescreen displays;
• Visual quality: Bright lighting, rich textures, and sharp details while maintaining the speed characteristic of the Flash line.

4. Create assets with full control over aspect ratios and resolutions from 512px to 4K.

From vertical social posts to 4K wide-screen backdrops, your visuals stay sharp, vibrant, and production-ready. pic.twitter.com/mKvw1AqozD

— Google (@Google) February 26, 2026

Nano Banana 2 will replace the Pro version in the Fast, Thinking, and Pro models of the Gemini app. Subscribers to Google AI Pro and Ultra will retain access to Nano Banana Pro.

The model is also integrated into Google Search (in AI and Lens modes) and is available in a preview version in AI Studio and the Gemini API.

Flow Update

Google has also introduced an updated version of the video generation service Flow. Developers have completely redesigned the interface, combining image and video generation capabilities into a single workspace.

Today we’re expanding Flow into a full AI creative studio. We’ve redesigned the experience and powerful new tools so you can draft, visualize, and refine your stories in a single, unbroken workflow.

What’s new:
✨ An updated interface for seamless workflows
🖼️ An easier way to… pic.twitter.com/xrUkNaSMiU

— Flow by Google (@FlowbyGoogle) February 25, 2026

The Flow service has integrated the best features from experimental platforms Whisk and ImageFX. Nano Banana 2 has become part of the core functionality. Users can now create images and immediately use them as frames for video generation without leaving the service.

The new asset grid simplifies the search, filtering, and sorting of images and videos, as well as grouping them into collections. The “Lasso” tool allows precise selection of image areas and modifications through text requests—such as “remove person” or “add koi fish to water.” Direct drawing on images to indicate changes is also available.

According to the company, since its launch, Flow users have created over 1.5 billion images and videos for creative projects, including films, music videos, and advertising campaigns.

Earlier in February, Google introduced Gemini 3.1 Pro and updated the AI tool for vibe coding Opal.