Google DeepMind released Nano Banana 2 Lite, an image generation model. This model, technically named Gemini 3.1 Flash Lite Image, is now available across the Google ecosystem, including Google AI Studio, the Gemini API, and Google’s Gemini Enterprise Agent Platform.

Nano Banana 2 Lite generates images in approximately four seconds when in its default low-thinking mode. In comparison, the standard Nano Banana model generates the same images in about 20 seconds. Google states the average cost for Nano Banana 2 Lite is $0.034 per 1,000 images.

The API rate for Nano Banana 2 Lite is $0.25 per 1 million input tokens and $1.50 per 1 million output tokens. This rate is half the rate of Nano Banana 2. For comparison, the API rate for Nano Banana Pro is $2 per 1 million input tokens and $12 per 1 million output tokens. Nano Banana 2 Lite is optimized for rapid-fire prototyping and can be accessed in Gemini by selecting the Flash-Lite option.

Google states Nano Banana 2 Lite serves as a replacement for the original Nano Banana model, which it now refers to as its legacy model. The original Nano Banana launched last summer and was powered by Gemini 3.1 Flash. Nano Banana 2 was released in February. Google indicated that building with generative media often involves creative iteration.

The company also stated that with these two models, developers can connect rapid image generation with video creation and editing to build comprehensive, end-to-end multimedia experiences. Images produced by Nano Banana 2 Lite contain SynthID watermarks. Google noted that Nano Banana 2 Lite has difficulty with very small text, infographics generated by the model are more likely to include incorrect data, and characters or people in outputs may show poor consistency across iterations. Arena.ai Elo scores indicate that users rate Nano Banana 2 Lite outputs nearly as highly as non-Lite versions.