To facilitate maximum utilization of “Nano Banana Pro,” Google provides a set of practical tips and guidelines that help users fully control the details of the images being created, while maintaining consistency and high quality. It is worth mentioning that the model is available globally within the Gemini application via the “Create image” option using the “Thinking” model, and all users can try it out at the moment.
To achieve the best results and increase control over images, Google advises adding the following elements when writing instructions and commands for the model:
* Subject: To accurately define who or what appears in the image, such as: “A robot with rigid features,” or “A person playing ball,” or “A tabby cat wearing a small wizard hat,” and so on.
* Composition: To determine the angle and style of photography, such as: “Very close shot,” or “Wide shot,” or “Low-angle shot,” or “Portrait.”
* Motion: To clarify the ongoing event within the image, such as: “Preparing a cup of coffee,” or “Driving a car fast,” or “Running through a field.”
* Location: To specify the location of the scene, such as: “A futuristic cafe on Mars,” or “A crowded chemist’s library,” or “An open garden full of trees.”
* Artistic Style: To determine the overall aesthetic character, such as: “Three-dimensional drawings,” or “Watercolor painting,” or “Realistic photography,” or “1990s product photography style.”
* Editing Instructions: When editing a generated image, the commands should be direct and precise, such as: “Change the color of the man’s tie to green,” or “Remove the car in the background.”
