Nano Banana 2.1 now available on AI Gateway
By Steven Van ·
The model supports product scene changes and edits confined to marked regions, with Flash-tier latency and cost.
Google’s Nano Banana 2.1 improves image generation and editing over earlier Nano Banana models, with a focus on placing products in new scenes, editing selected regions and accurately depicting real-world subjects. It is available through Vercel AI Gateway.
Product scenes and targeted edits
Product recontextualization places an existing product in a new scene. Mask- and ink-based editing changes only the area marked by a mask or strokes drawn on the image. Vercel says the model renders photorealistic skin tones, intricate materials, sharp lighting and coherent backgrounds at the latency and cost of a Flash-tier model.
Call it from code or the terminal
Use the model name google/gemini-nano-banana-2.1 through the AI SDK, the OpenAI-compatible Chat Completions API or AI CLI. In the AI SDK example, generated images are returned in result.files. AI CLI accepts the model ID through its -m option, and the model is also available to try in the model playground.
The release links to AI Gateway API key documentation for authenticating requests.