Google has launched Nano Banana 2, a new visual AI model, and updated its Flow service to enhance image and video generation capabilities.
Google has introduced a significant upgrade to its AI capabilities with the launch of Nano Banana 2, a visual generation model designed to operate at the speed of Gemini Flash.
This new model, which aims to enhance image generation and editing, is now available across the Gemini app, Google Search, and various developer tools. Google touted it as their most advanced image generation model to date, stating, "Pro-level quality, at Flash speed. Rolling out today across @GeminiApp, Search, and our developer and creativity tools."
Nano Banana 2 utilizes the expansive Gemini knowledge base and real-time data from web searches, allowing it to accurately render intricate objects, create infographics, and convert text notes into visual diagrams. Notably, it can produce clear text within images, catering to various formats like marketing materials and greeting cards.
The model boasts several improvements over its predecessor, including:
- Object consistency: The ability to maintain the appearance of up to five characters and accuracy for 14 objects in one project.
- Execution accuracy: Enhanced understanding of complex prompts and attention to detail, closely following user instructions.
- Format flexibility: Users can control aspect ratios and resolutions, ranging from 512px to 4K, making it versatile for different display needs.
- Visual quality: Produces bright, textured images with sharp details while retaining the speed characteristic of the Flash line.
Nano Banana 2 will replace the Pro version in the Fast, Thinking, and Pro models of the Gemini app, while subscribers to Google AI Pro and Ultra will still have access to Nano Banana Pro. The model is also integrated into Google Search, available in AI and Lens modes, and can be accessed in beta through AI Studio and the Gemini API.
Flow Service Update
In addition to Nano Banana 2, Google has revamped its video generation service, Flow. The updated interface now combines image and video generation into a single workspace, enhancing the user experience. Google stated, "Today we're expanding Flow into a full AI creative studio," highlighting new tools for drafting and visualizing creative projects seamlessly.
The Flow service now incorporates features from the experimental platforms Whisk and ImageFX, with Nano Banana 2 integrated into its core functionality. This allows users to create images and utilize them as video frames without leaving the platform. New features like an asset grid simplify the process of searching and organizing images and videos, while the Lasso tool enables precise modifications based on text prompts. Users can also directly draw on images to indicate changes.
Since its launch, Flow users have created over 1.5 billion images and videos for various creative endeavors, including films and advertising campaigns. This follows Google's earlier release of Gemini 3.1 Pro and the update to its vibe coding tool, Opal, in February.
