Google launches Nano Banana 2, a faster AI image generator powered by Gemini

Google is doubling down on AI image creation. On Thursday, the company introduced Nano Banana 2, the latest version of its widely used image generator, signaling a push to make high-quality visuals faster and easier to produce across Gemini.

The original Nano Banana launched in August and quickly gained traction online. Google followed with Nano Banana Pro in November, built on Gemini 3 Pro and aimed at higher-fidelity workloads. The new release sits between speed and accuracy, bringing many Pro-level capabilities to a broader group of users.

According to Google, Nano Banana 2 draws on Gemini’s real-time knowledge to produce more accurate visual outputs. The model is built to deliver quicker results, stronger instruction adherence, and sharper text rendering for use cases such as marketing mockups and greeting cards.

“Introducing Nano Banana 2 (Gemini 3.1 Flash Image), our latest state-of-the-art image model. Now you can get the advanced world knowledge, quality and reasoning you love in Nano Banana Pro, at lightning-fast speed,” Google said in a blog post.

Nano Banana 2 Is Here: What Changed in Google’s Wildly Popular AI Image Generator.

Courtesy : Google

The company said Nano Banana Pro will remain available for “high-fidelity tasks requiring maximum factual accuracy,” whereas Nano Banana 2 is tuned for “rapid generation, precise instruction following, and integrated image-search grounding.” With this update, Nano Banana 2 replaces the earlier model across Gemini’s Fast, Thinking, and Pro tiers.

The release lands as AI image and video tools continue to gain momentum with consumers and businesses. Text prompts now produce increasingly detailed visuals and short-form video, turning generative media into one of the most competitive fronts in AI. OpenAI entered the video race with Sora in 2024, and CEO Sam Altman wrote last March that heavy demand was “melting” the company’s AI processors.

Adobe has been moving along a similar path, weaving generative capabilities deeper into its creative stack through Firefly, its image and video generation platform.

At the same time, the growth of generative media continues to stir debate over copyright and ownership. Creative companies have raised concerns that these systems may rely on protected material. ByteDance has already faced pushback from major Hollywood studios, including Disney and Paramount, over alleged intellectual property issues tied to its Seedance video tool.

Google says the update brings Gemini Flash capabilities to visual generation, enabling faster edits and broader access to features previously limited to Pro users.

The company highlighted two core improvements:

Advanced world knowledge. The model pulls from Gemini’s real-time knowledge and web search to render subjects more accurately. Google says this depth helps with tasks such as infographic creation, diagramming, and data visualization.

Precision text rendering and translation. Nano Banana 2 is built to generate clearer, more legible text for marketing mockups and greeting cards. The model can translate and localize text directly inside images for global use.