Key facts
- Google DeepMind has launched Nano Banana 2 Lite, a new AI image generation model.
- The model is part of the Gemini 3.1 Flash Lite Image family.
- Nano Banana 2 Lite generates images in approximately 4 seconds.
Google DeepMind has launched Nano Banana 2 Lite, a new AI image generation model designed for speed and affordability. The model can produce images in approximately 4 seconds at a cost of $0.034 per 1,000 images, making it suitable for high-volume workflows.

The release of Nano Banana 2 Lite offers developers a faster and more cost-effective tool for AI-generated imagery, potentially accelerating content creation workflows and innovation in multimedia experiences.
Google DeepMind has launched Nano Banana 2 Lite, a new AI image generation model designed to balance quality, speed, and cost-effectiveness. This model, technically Gemini 3.1 Flash Lite Image, is optimized for high-volume workflows and rapid idea exploration, generating images in approximately 4 seconds, a significant improvement over the standard Nano Banana's 20-second generation time. Google claims the model is also more affordable, with API access costing $0.034 per 1,000 images, and input and output tokens priced at $0.25 per 1 million and $1.50 per 1 million, respectively. These rates are half those of the standard Nano Banana model. However, Google acknowledges limitations, including potential difficulties with rendering small text, inaccuracies in infographics, and inconsistent character representation across multiple image generations. Nano Banana 2 Lite is now available via Google AI Studio, the Gemini API, and Google’s Gemini Enterprise Agent Platform, serving as a replacement for the company's legacy model.
Pick the topics you care about. Get only what matters, on your cadence.