infinix x yugatech

Google rolls out Gemini 2.5 ‘Nano Banana’ for advanced AI image editing

Listen to article

Google has introduced Gemini 2.5 Flash Image, also called ‘Nano Banana,’ in the Gemini app. Developed by Google DeepMind, the model enhances AI-powered photo editing with improved likeness preservation.

The new feature focuses on keeping people and pets consistent when changing backgrounds, outfits, or time periods. This tool is meant to help users avoid unrealistic or noticeable distorted edits.

Here’s a sample photo from Google’s article about their state of the art image model below.

Gemini 2.5 also adds multi-step editing, allowing users to apply changes one at a time. A style transfer option is included, letting users add patterns or textures from one photo to another.

All AI-edited images will include a visible watermark and an invisible SynthID tag for authentication. Interested users may check it out now as the update is now available via the Gemini API, Google AI Studio, and Vertex AI for enterprise.

Readers should note that Gemini 2.5 Flash Image is priced at USD 30 per 1 million output tokens. Each image is worth 1,290 output tokens or USD 0.039 per image. Other modalities on input and output follow the Gemini 2.5 Flash pricing.

React to this article:
Written by
Duane Villanueva

Communication graduate, closet cynic, and kid at heart. Duane is a rare person to find, quite literally. He often takes to himself but has proven his mettle in tech media with his quick wits. Well, the portfolio of scriptwriting, web content, and public relations help too, we suppose. As a homebody, he often spends his time on the streaming platform Twitch or ‘farming’ gaming clips with friends. He is also an avid fan of round glasses and anything relative to blueberries.

View all posts by Duane Villanueva →

1 Comment

HA
Harold Cabradilla Casuga · 10 months ago

Edit picture

Reply

Leave a Reply

Loading next article...