Google is gearing up for a new era of AI image generation with the upcoming Nano Banana 2 model. This follow-up to the original Nano Banana model will be part of the Gemini app and promises smarter, self-correcting image creation.
The new model, internally known as GEMPIX 2, will adopt a multi-step workflow where it plans the image before making it, analyzes it for mistakes, fixes them, and repeats the process until it’s ready. This built-in self-correction is new for Google’s image tools, treating AI image generation like a real design assistant.
The leaked previews show cleaner lines, sharper angles, and fewer errors in AI-generated images. Nano Banana 2 will bring better angle and viewpoint control, more precision in coloring, and the option to fix text in the image without messing with the rest of the output.
This model is expected to be more than just a resolution bump, as it will change how Google thinks about AI image generation. The multi-surface rollout will follow the same playbook as the original Banana drop, making AI-generated images look better.
With Nano Banana 2, Google is teaching its model to notice its own mistakes and fix them proactively, mirroring the human creative process more than past AI tools have. This shift marks a philosophical change in how AI image generation works, offering users a more realistic and reliable experience.
Source: https://www.techradar.com/ai-platforms-assistants/gemini/google-will-peel-back-a-new-era-of-ai-images-with-nano-banana-2