Google: Nano Banana Pro (Gemini 3 Pro Image Preview)

Google: Nano Banana Pro (Gemini 3 Pro Image Preview) is an image generation and editing model from Google, built on Gemini 3 Pro. It is the successor to the original Nano Banana, and Google describes it as its most advanced image generation and editing model, with improved multimodal reasoning and real-world grounding compared with the earlier version. It sits in the image category while also handling text, so it is a multimodal model rather than a pure image generator.

The model accepts image and text inputs and can return both image and text outputs. It has a context window of 65,536 tokens and a maximum output of 32,768 tokens. Generated images can be requested at 1K, 2K or 4K resolution, and you can choose from a set of aspect ratios: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9 and 21:9. The n parameter is fixed at 1, so each request produces a single image, and you can supply up to 14 input references alongside your prompt.

Those capabilities point to a few practical uses. Because the model takes image inputs as well as text, it fits editing and transformation work where you supply one or more reference images and describe the change you want. The reference limit of 14 supports workflows that combine several source images, such as compositing or style matching across a set. The resolution and aspect ratio options make it usable for both square social formats and wide or tall layouts, and the text output means a single call can return an image together with a written response. The reasoning and grounding features suit prompts that need more than surface-level pattern matching, for example scenes with specific objects, spatial relationships or real-world context.

You call it through api14's OpenAI-compatible API using the model id "google/gemini-3-pro-image-preview". The supported request parameters are include_reasoning, max_tokens, reasoning, response_format, seed, stop, structured_outputs, temperature and top_p. Image-specific options such as resolution, aspect ratio and input references are set through the request as well. Because the interface is OpenAI-compatible, you can point an existing OpenAI client at api14 and change the model id rather than rewriting your integration.

Keep the limits in mind when choosing this model. The context window is 65,536 tokens and the maximum output is 32,768 tokens, and n is capped at 1, so you cannot request multiple images in a single call. If your task is text-only, or you need several images per request, another model may fit better.