Google has launched the next-generation version of its image creation and editing AI tool, “Nanobanana,” called “Nanobanana2.” The new version is characterized by significantly increased generation speed while maintaining the high-quality performance of the previous “Pro” level.
On the 26th (local time), Google announced that it would introduce “Nanobanana2,” which is based on the “Gemini 3.1 Flash Image.” According to Google, this combines the advanced reasoning capabilities revealed with “Nanobanana Pro” last November with faster processing speeds.
As a result, users can now enjoy the high-quality images that were previously part of the paid “Pro” features at a faster speed and for free.
The tool also supports translating and localizing text within images into other languages. It can integrate real-time web searches and Gemini’s global knowledge to produce visual materials that reflect the latest information, making it highly useful for infographics or data visualization tasks.
It can maintain consistency with up to five characters and 14 object elements in a single task, reducing issues of character appearance changes in book illustrations or comic production.
The resolution ranges from 4K Ultra HD to 512-pixel low-resolution, and it offers various aspect ratios such as 4:1 and 1:8, providing creators with more options. For professional users, a function to adjust the model’s reasoning level has been added to produce more sophisticated results in complex tasks.
Nanobanana2 will be implemented as a default model in major Google products, including the Google Gemini app, from today.
Meanwhile, since its first release in August last year, Nanobanana attracted 13 million new users to the Gemini app in just four days and generated over 5 billion images by mid-October, achieving significant success.