Washington, USA: Google has launched Nano Banana 2.1, the latest iteration of its AI image generation and editing model, bringing sharper visual quality, more precise editing and stronger consistency for characters and objects across multiple images and successive edits.
The model builds on Nano Banana 2, also known as Gemini 3.1 Flash Image, and is based on Gemini 3.6 Flash. Google says the upgrade is designed to deliver more capable image creation and editing while retaining the speed and efficiency of its Flash architecture.
Nano Banana 2.1 delivers improvements across several core areas, including visual design, mask-based editing and subject consistency. Google says the model produces more sophisticated visual compositions and more natural-looking images, while offering greater control over targeted edits.
Its enhanced mask-based editing capabilities allow users to isolate and modify specific portions of an image without altering the rest of the scene, enabling more precise changes and faster iteration.
The model also strengthens consistency across characters and objects, allowing subjects to retain their recognizable appearance across multiple images and successive edits. Google says Nano Banana 2.1 can maintain the resemblance of multiple characters and the fidelity of multiple objects within a single workflow, making it better suited to visual storytelling, product development and other applications requiring continuity across images.
Beyond editing, the model improves text rendering and infographic design and can draw on Google’s real-world knowledge and search capabilities to produce more accurate visual representations of specific subjects. It also supports image upscaling to 2K and 4K resolutions and flexible aspect ratios for different output formats.
Google is rolling out Nano Banana 2.1 across its ecosystem, including the Gemini app, Google Search’s AI Mode, Google AI Studio, Google Flow, Google Stitch, Google Ads and Gemini Enterprise. The model is also available to developers through the Gemini API under the model identifier gemini-nano-banana-2.1.
Nano Banana 2.1 supports 1K, 2K and 4K image generation, with the latest developer documentation also listing text, image, video and PDF as supported input types.
