Google unveiled the new model of artificial intelligence Nano Banana 2 with faster speed

Google by introducing a new image production model Nano Banana 2 (with the official name Gemini 3.1 Flash Image), improved the speed and quality of creating artificial intelligence images to a higher level.

Google unveiled the latest version of its artificial intelligence model for image production and editing called Nano Banana 2. Officially known as Gemini 3.1 Flash Image, this model can create more realistic and detailed images than the previous version, while operating faster.

Combining speed and quality

Nano Banana 2 retains some of the high quality and image fidelity features of the previous model (Nano Banana Pro) but produces images at a much faster rate. This model has the ability to create images in resolutions from 512 pixels to 4K and in different ratios.

The important features of this model are:

• The ability to maintain the consistency of the faces of up to five different characters and the accuracy of displaying up to 14 objects in one image (for better storytelling).
• Produce images with more vivid lighting, richer details and textures, and greater accuracy in response to complex user requests.

Extension to Google products

Google has announced that Nano Banana 2 will now be the default image generation model in the Gemini app in Fast, Thinking and Pro modes, and has also been added to other Google tools such as Google Lens and AI mode in search.

Users on the Google AI Pro and Ultra plans can still use the previous version of Nano Banana Pro for certain tasks.

read more

The original version of Nano Banana was released in August 2025 and quickly became popular among users, then the upgraded Pro version was released in November 2025. With the launch of Nano Banana 2, Google is trying to bring the speed, quality and availability of AI image generation tools to a wider range of users.