Washington, July 01 (QNA) - Google announced the launch of Nano Banana 2 Lite, an artificial intelligence image generation model optimized for speed and lower costs, targeting businesses and developers that need to generate large volumes of images for commercial applications. 
The model, (also called Gemini 3.1 Flash-Lite Image) is Google's fastest and most cost-efficient image generation model. According to Google, it can generate a 1K-resolution image in under four seconds at a cost of just $0.034 per 1K image.
The model is now available through Google AI Studio, the Gemini API, and the Gemini Enterprise Agent Platform, targeting enterprise and business developers.
Google said in a blog post that Nano Banana 2 Lite is built on the Gemini 3.1 Flash-Lite architecture and was designed to address one of the key challenges in image generation models: balancing execution speed with computational resource consumption.
The model is intended for environments that require high-volume image production, such as advertising platforms, e-commerce, and rapid development tools, while reducing response times and operating costs compared with larger models.
The new release represents a significant improvement over its predecessor, Nano Banana, which was based on Gemini 2.5 Flash Image. Enhancements include improved general knowledge, better scene and contextual layout generation, greater consistency of characters and objects across a sequence of images, and improved multilingual text rendering within images.
Nano Banana 2 Lite is limited to generating images at 1K resolution, unlike Nano Banana 2 and Nano Banana Pro, which support resolutions of up to 4K. However, Google said this limitation contributes to greater efficiency.
Nano Banana 2 Lite is aimed at software developers, advertising platforms, and e-commerce companies, with a focus on automated content creation while maintaining consistency of elements across images.
Google also noted that image editing may take slightly longer than image generation because of the additional processing involved. (QNA)