Google has released Nano Banana 2.1, an updated image generation and editing model designed for precise object manipulation. The update introduces mask based editing and improved character consistency across sequential generations up to 4K resolution. The model is now accessible across Gemini services, Google Search AI Mode, Google Ads, and developer platforms.
The primary addition to Nano Banana 2.1 is mask based editing, allowing users to select and modify specific image areas without rebuilding the entire scene. The system also improves character and object consistency across multiple outputs. This adjustment helps maintain the appearance of individuals or items across different visual scenes without variation.
Google states the model delivers higher realism, cleaner text rendering, and better handling of infographics compared to the previous version. The system supports multi image references to guide new outputs and generates visuals in resolutions up to 4K.
The model is live in the Gemini application, Google Search AI Mode, Google AI Studio, Flow, Stitch, Google Ads, and the Gemini Enterprise platform. Developers can access the model through the API using the gemini nano banana 2.1 identifier.
