Explore curated image generation and editing prompts for Nano Banana, then open them directly in the AI Raphael workspace.
Model Guide
Nano Banana is a Google AI image model family for generation and editing, covering Nano Banana Pro and Nano Banana 2. It is designed for natural-language photo editing, character consistency, scene preservation, and high-resolution visual output.
The model understands edit intent in plain language and turns local changes, background replacement, composition updates, and lighting changes into visual results.
Useful when identity, expression, proportions, camera angle, and spatial relationships need to stay coherent across edits or image series.
Strong for product visuals, posters, social assets, and realistic scenes where textures, reflections, shadows, and fine details matter.
Useful for headlines, signage, packaging, and multi-subject scenes where the image structure needs to stay stable.
The core value of the Nano Banana family is turning image editing from complex tool operations into natural-language instructions. It can analyze people, objects, backgrounds, lighting, textures, and spatial relationships while keeping edited images natural.
Nano Banana Pro is stronger for high-quality editing, character consistency, multi-element scenes, and 2K-4K-ready output. Nano Banana 2 is better for fast enhancement, realistic detail, text rendering, and everyday visual production. Both models work well in generation plus editing workflows.
For photo-like edits, product visual remakes, brand visual extensions, or consistent character series, Nano Banana is often a better fit than a one-shot text-to-image model because it emphasizes visual semantics, local edits, and source-image continuity.
Answers about Nano Banana capabilities, use cases, and how it differs from GPT Image 2.