Access frontier-level AI performance for image tasks with low latency
As part of the Gemini 3.5 series, Flash variant delivers optimized performance for agents and real-time applications requiring fast image processing
Gemini 2.5 Flash Image
Gemini 2.5 Flash Image
Meet "nano-banana🍌" the new SOTA image model
Overview
What it is
Search the world's information, including webpages, images, videos and more. Google has many special features to help you find exactly what you're looking for.
Intent
I need it when
Generate or edit images from text descriptions and visual inputs
Gemini 2.5 Flash Image is a multimodal AI model that can create and process images from various input types, enabling users to produce visual content programmatically without manual design work
Analyze and understand image content for research or automation
The model processes images as input alongside text, allowing users to extract insights, classify visual data, or automate image-based workflows at scale
Build AI applications that combine vision and language capabilities
Gemini 2.5 Flash Image integrates with Google AI Studio and developer tools, enabling developers to create agents and applications that reason about and generate visual content
Drop
Not a fit when
- User requires detailed pricing information or transparent cost structure for budget planning
- User needs a product with published service level agreements or guaranteed uptime commitments
- User requires offline-only image processing without cloud dependencies
- User needs a lightweight solution with minimal computational requirements
- User requires open-source code or full transparency into model architecture
Commercials
Pricing
Pricing not specified