Google has launched the Nano Banana 2.1 model with reduced costs and enhanced capabilities for image generation and editing.
Google has introduced Nano Banana 2.1, its latest model for image generation and editing, part of the Gemini 3 series and based on the advanced Gemini 3.6 Flash architecture.
Significantly, the cost for generating images through the new model has been reduced to $30 per 1 million tokens, a drop from the previous $60 cost.
The model boasts an impressive capacity to process contexts up to 1 million tokens, delivering text outputs of up to 64,000 tokens while generating images in native resolutions up to 4K.
Key improvements include enhanced text display on posters and charts, greater accuracy in infographic representation, and consistent character appearance during batch editing. Users can also leverage a masking feature that allows for targeted modifications of specific image areas based on text or sketches.
Already, Google is incorporating Nano Banana 2.1 into various platforms, including the Gemini app, the AI Mode in its search function, Google Ads, and Google AI Studio. The earlier version, Nano Banana 2, will be phased out on October 29.
Moreover, Google recently unveiled Gemini 4 Argon, a cutting-edge AI model focused on programming and cybersecurity solutions.