Google Nano Banana 2.1 Brings GA Image Generation and Editing to Gemini

Google has officially released Gemini Nano Banana 2.1, a new model for image generation and editing, now generally available on the Gemini Enterprise Agent Platform. Designed for multimodal workflows, the model supports interleaved text and image inputs, allowing users to combine prompts with source imagery. It offers output resolutions up to 4K and includes content credentials for provenance. While it provides robust capabilities for marketing and product visualization, the model does not support certain standard parameters like temperature or seed, nor does it currently support fine-tuning or RAG. Pricing is based on a pay-as-you-go, token-consumption model, with costs scaling according to the requested resolution. Organizations are encouraged to validate the model's performance and output quality against their specific operational requirements before integrating it into production environments, as real-world latency and throughput figures remain to be tested.