Google, one of the leading tech brands, has unveiled its latest AI image generation and editing model, Gemini 2.5 Flash Image, also nicknamed Nano Banana. This upgraded model enhances creativity and consistency, offering users smarter editing features directly in the Gemini app. Available worldwide for both free and paid users, the new model enables advanced edits while ensuring that people, pets, and objects maintain the same look across multiple images.
Gemini 2.5 Flash Image: Key features
The new Nano Banana model comes with multiple improvements:
- Character consistency: Keeps the same identity across different edits and images.
- Prompt-based editing: Make edits using natural language commands like changing backgrounds or removing objects.
- Multi-image fusion: Blend multiple photos into a single scene with new styles, textures, and colours.
- Multi-turn editing: Apply step-by-step edits such as adding furniture, changing hairstyles, or retouching specific parts of a photo.
- Style transfer: Mix and match designs, colours, and patterns across different images.
All generated or edited images carry Google’s visible watermark and invisible SynthID watermark, ensuring transparency.
Integration with Gemini App
With this update, the Gemini app now allows users to perform complex edits such as:
- Blending group portraits and pet photos into one frame.
- Retouching black-and-white photos by adding colours.
- Editing step-by-step while preserving the main subject.
These capabilities make the app more versatile for personal use, storytelling, brand asset creation, and product visuals.
For developers and enterprises
Developers can access Gemini 2.5 Flash Image through Google AI Studio, Gemini API, and Vertex AI. Google has also released template apps to demonstrate features like multi-image fusion and brand asset generation.
The pricing is set at USD 30 per 1 million output tokens, with each generated image costing around USD 0.039.