Google Launches ‘Nano Banana’ AI Photo Editing Tool

Google officially unveiled its next-generation AI image editing tool, originally codenamed “Nano Banana” and now released as Gemini 2.5 Flash Image. Developed by Google DeepMind, the tool brings unprecedented ease and creativity to image editing, enabling users to transform, remix, and merge photos using just natural language prompts.

Google CEO Sundar Pichai hinted at the launch on social platform X with a quirky post featuring banana emojis and fun AI-edited images of his dog, Jeffree, in celebration of International Dog Day.

What Is Nano Banana?

Nano Banana is the code name for Gemini 2.5 Flash Image, an AI-powered tool that allows users to,

• Edit photos using natural language
• Maintain consistency of people, pets, or objects across edits
• Merge multiple images into a single, coherent scene
• Apply new fashion styles, change hairstyles, or alter settings
• Add creative elements like superhero costumes or themed backgrounds

The tool is available through the Gemini app, Google AI Studio, and Vertex AI, making it accessible for both everyday users and enterprise developers.

Unique Features of Nano Banana

1. Character Consistency

One of its core strengths is maintaining identity and visual continuity across different images. Whether it’s a person, pet, or object, the tool ensures that likeness is preserved even after multiple edits or scene changes.

2. Image Merging

Users can join separate photos into one, placing themselves and pets or friends in the same frame—even if the original pictures were taken at different times or places.

3. Design Mixing

An innovative feature called “Design Mixing” allows users to transfer styles or patterns from one image to another. For example, applying a floral design to an outfit or mimicking a texture from another object.

4. Prompt-Based Editing

Users simply upload an image and type what they want—like “change background to beach” or “add superhero costume”—and the tool does the rest.

Pricing and Access

• Available via Gemini API, Google AI Studio, and Vertex AI
• Cost per image: $0.039 (1290 output tokens)
• Full model pricing: $30 per 1 million output tokens
• This makes the tool both high-performing and affordable, ideal for individuals, developers, and businesses.