Why Google Chose the Strange Name ‘Gemini Nano Banana’ for Its New AI Model

Google's latest image generation model, Gemini Nano Banana, has captured global attention not just for its impressive capabilities but also for its peculiar name. Until recently, few knew the true story behind how this unusual title came to be.

Now, David Sharon, Group Product Manager for the Gemini App, has shared the full story on the Made by Google podcast, and it is as unexpected as the name itself.

A Name Born Out of Anonymity

According to Sharon, the name "Nano Banana" was never meant to be permanent. It started as a simple placeholder used when the model was first submitted anonymously to LM Arena, a public platform where AI models are tested without revealing their creators.

"Nano Banana was created by a PM named Nina," Sharon explained. "When you submit a model anonymously to LM Arena, you must include a placeholder name."

Nina, working late one night around 2:30 a.m., came up with the quirky name "Nano Banana" on a whim. To her surprise and later to Google's, the placeholder name caught on quickly. "People were already calling it Nano Banana, so we decided to keep the name," Sharon said.

From Placeholder to Public Favorite

What began as a random internal label quickly became a popular identity among users. Since the model was submitted anonymously, people interacted with it based only on its performance, without knowing that it was developed by Google.

As positive feedback started pouring in, "Nano Banana" transformed from a temporary nickname into a lasting identity. The light-hearted and playful tone of the name connected with users, making the model stand out among other AI tools on the platform.

Pushing Boundaries in Image Generation

Beyond its funny name, Gemini Nano Banana represents a major advancement in AI image generation. Sharon said the model focuses on consistency in subject identity, which means it can accurately reproduce people, pets, or objects across multiple images, even in creative or imaginative settings.

"When I first tested Nano Banana, I uploaded an image of myself and asked to place myself in space. For the first time, I saw myself, not a distorted version," Sharon shared. This marked an important breakthrough in generating realistic and recognizable visuals.

The development team prioritized personalization and realism so that users could see familiar faces, including their own or their pets, rendered clearly in artistic and fantasy scenes.

The Creative Edge: Conceptual Blending and Innovation

An internal "Greenfield" team at Google tested Nano Banana through a series of creative experiments. They explored its ability for conceptual blending, where the model merges unrelated ideas into a single, meaningful image.

In one such test, the team asked the model to combine a couch and a potato. The result was a clever and humorous image of a literal "couch potato."

This showed that Nano Banana could go beyond traditional image generation. It could create visuals that are both realistic and imaginative, demonstrating how far AI has come in understanding and expressing abstract human ideas.

A Playful Name for a Serious Innovation

In the end, the name Gemini Nano Banana serves as a reminder that innovation does not always come from serious moments. What started as a random, late-night placeholder became a symbol of creativity and spontaneity within Google's AI team.

By choosing to keep the name, Google embraced the humor and originality behind its creation. The story of Nano Banana highlights how even the most advanced technology can carry a lighthearted human touch.