Here are the details on Google's Gemini 2.5 Flash Image model, nicknamed 'Nano-Banana,' and its revolutionary features!
In a move set to transform image creation, Google has launched the Gemini 2.5 Flash Image model, affectionately nicknamed “Nano-Banana.” This new model, now integrated into the Gemini app for both free and paid users, brings a suite of advanced image generation and editing capabilities directly to a global audience.
Nano-Banana addresses a key user demand: more creative control. Following the success of native image generation in Gemini 2.0 Flash, users had asked for higher-quality images and greater flexibility. The new model delivers on this, offering features that allow for unprecedented consistency and precision.
Key features of the model include:
- Character Consistency: Nano-Banana can maintain the consistent appearance of a person, pet, or object across multiple images and edits
- Prompt-Based Editing: Say goodbye to complex software. With Nano-Banana, you can simply use natural language commands to make precise edits
- Multi-Image Fusion: Imagine placing a new object into a scene, restyling an environment with different textures, or even combining portraits of people and pets in one seamless image
- Enhanced World Knowledge: Leveraging the core intelligence of Gemini, the model can interpret complex diagrams, answer questions, and follow intricate editing instructions
Availability, Pricing, and Transparency
The updated image editing capabilities are available to users worldwide via the Gemini app. For developers and enterprises, the Nano-Banana model can be accessed through the Google AI Studio, Gemini API, and Vertex AI.
Google has set a clear pricing structure, with the model priced at $30 per 1 million output tokens. Each generated image costs approximately 1,290 output tokens, which translates to about $0.039 per image. Other input and output modalities will follow the standard Gemini 2.5 Flash pricing.
In a commitment to transparency, all images generated or edited using the Gemini app will feature both a visible watermark and an invisible SynthID watermark. This measure ensures that AI-created content is clearly identifiable.
The launch of Nano-Banana marks a significant step forward in Google’s mission to make sophisticated AI tools more accessible and powerful for everyone, from everyday users to creative professionals.
