Google DeepMind has introduced Nano Banana 2 Lite, its quickest and most affordable AI image generation model to date. The tool, officially called Gemini 3.1 Flash Lite Image, produces pictures in seconds and costs significantly less than prior versions.
The model belongs to the Gemini 3.1 family and became available on June 30. It targets rapid prototyping where speed matters more than perfect detail, generating images in roughly four seconds compared with 20 seconds for the standard Nano Banana version.
API pricing starts at $0.034 per 1,000 images, half the rate of Nano Banana 2. Input tokens cost $0.25 per million and output tokens $1.50 per million. All outputs carry SynthID watermarks to identify them as AI-generated.
Users can access the model through Google AI Studio, the Gemini API, or the Gemini app by selecting the Flash-Lite option. Google also expanded availability of Gemini Omni Flash for video tasks in the same platforms.
The company notes that the model may struggle with small text, data accuracy in infographics, and character consistency across multiple images.
