"Google's new version of Nano Banana is now online."

On November 20th local time, Google's next-generation multimodal AI model, the second-generation Nano Banana update, was officially released with the codename Gemini 3 Pro Image Preview. Unlike traditional image models, Gemini 3 Pro Image is an inference model that performs internal reasoning before generating images, similar to thinking about how to best complete a task. This results in significantly improved image quality, higher accuracy, and better multi-language long-text rendering capabilities. According to the official blog, Gemini 3 Pro Image is designed to solve the most challenging image generation tasks. The model is now available on the Vertex AI platform and can also be accessed through Google's Gen AI SDK in Google AI Studio.