Google has launched its latest image generation and editing model, Nano Banana 2.1, which is part of the Gemini 3 family and built on the Gemini 3.6 Flash architecture.
The cost of generating images via the API has been halved compared to its predecessor.
Nano Banana 2.1 can handle input contexts of up to 1 million tokens. The model can process both text and images, providing text outputs of up to 64,000 tokens and images with native resolutions reaching 4K.
Enhancements include improved text rendering on posters and charts, better accuracy in displaying infographics, and maintaining character appearance during batch edits. The model supports mask functionality, enabling users to modify specific areas of an image based on text prompts or sketches without affecting the rest of the scene.
In the Gemini API, the cost for image generation is set at $30 per 1 million tokens, down from $60 for the previous version. Generating a single image ranges from $0.0336 for 1K resolution to $0.0756 for 4K resolution.
Google has already begun integrating Nano Banana 2.1 into the Gemini app, AI Mode in the search feature, Google Ads, as well as Google AI Studio, Flow, and Stitch platforms. The previous version, Nano Banana 2, will be discontinued on October 29.
Additionally, on September 30, Google introduced Gemini 4 Argon, a new flagship AI model aimed at programming, enterprise solutions, and cybersecurity.
Follow ForkLog on social media:
Telegram (main channel) Facebook X