Combine multiple reference images with a text prompt to generate unique AI creations.
Your AI-generated image will appear here
The Nano Banana Image Editor leverages a powerful multi-modal AI model that can understand both images and text simultaneously. This allows you to provide it with a rich context for your creations. Instead of just describing what you want, you can show it.
one or more reference images using the Multiple Images upload field — you can combine people, products, backgrounds, or art styles.
each uploaded image using the Label field (e.g., "person", "background", "cat", "art style") so you can reference them in your prompt.
in the Hint field per image for specific details (e.g., "red dress", "sunset lighting", "oil painting texture").
a style from Popular Style Instructions or write a custom Prompt describing what you want the AI to create.
in the prompt — e.g., "Put the person from image 1 in front of the background from image 2, in the style of image 3."
your Output Shape based on where you plan to use the image.
upload a Style Reference Image for additional style guidance.
and download your AI-created composite.
— using labels like "person", "background", "style reference" helps you write precise prompts and helps the AI understand each image's role.
— "smiling", "soft lighting", "watercolor texture" gives the AI granular control over how to interpret each reference.
— this tool excels at merging subjects from different images. Upload a person + a background + a style reference for best results.
— "Draw the person from image 1 standing on the beach from image 2" is far better than "combine the images."
— the AI cannot add detail that doesn't exist in your references.
— they're professionally crafted prompts that you can customize further in the Prompt field.
— extra images that aren't referenced in your prompt can confuse the AI and dilute the result.
