Google launches Nano Banana 2 for faster image generation, more: What's new

Google launches Nano Banana 2 based on Gemini 3.1 Flash Image, combining faster image generation speeds with improved reasoning, better text rendering and stronger visual quality

Google has announced Nano Banana 2, the new version of the company’s image generation and editing tool. Nano Banana 2 is powered by the Gemini 3.1 Flash Image model and is said to combine advanced knowledge, improved image quality and stronger reasoning with faster generation speeds.

 

The original Nano Banana was introduced in August last year for image creation and editing. In November, Google followed it up with Nano Banana Pro, which focused on higher intelligence and more detailed creative control. According to Google, Nano Banana 2 is designed to bring together the key strengths of both versions.

Nano Banana 2: What’s new

In the blog, Google said Nano Banana 2 brings the speed of Gemini Flash to image generation, enabling faster edits and quicker iteration. It added that the model also makes several features that were previously limited to the Pro version available to a wider set of users.

 

• Advanced world knowledge: The model uses Gemini’s knowledge base along with real-time information and images from web search to better represent specific subjects. Google said this also helps users create infographics, convert notes into diagrams and generate data visualisations.

• Text rendering and translation: Nano Banana 2 can generate clearer and more accurate text within images for uses such as marketing mockups or greeting cards, Google said. It can also translate and localise text directly inside an image.

Improved creative control

According to Google, Nano Banana 2 reduces the gap between speed and visual quality, delivering high-quality, photorealistic images while maintaining fast performance. Here’s what the new model offers and how it improves on the original Nano Banana:

• Better balance of speed and quality: Nano Banana 2 aims to deliver high-quality, photorealistic images while maintaining fast generation speeds.
• Subject consistency: It can maintain the resemblance of up to five characters and preserve up to 14 objects in a single workflow, helping users create storyboards without changing how elements look.
• Improved instruction following: The model is designed to follow complex prompts more accurately, so the final image more closely matches what the user requested.
• Flexible output settings: Users can choose different aspect ratios and resolutions, from 512px up to 4K, depending on where the image will be used.
• Upgraded visual quality: The model produces improved lighting, richer textures and sharper details compared to the original Nano Banana.

Availability

Nano Banana 2 is rolling out across several Google products. In the Gemini app, it replaces Nano Banana Pro across Fast, Thinking and Pro modes, though Pro subscribers can still access Nano Banana Pro when needed.

 

The model is also available in Search through AI Mode and Lens, AI Studio and the Gemini API, Google Cloud’s Vertex AI, Flow and Google Ads. Google said Nano Banana 2 is designed for fast generation and precise instruction following, while Nano Banana Pro remains available for high-fidelity tasks requiring maximum factual accuracy.