Nano Banana Prompt Guide: Examples and Best Practices for AI Image Generation

While sharing a name, Nano Banana and Nano Banana Pro are fundamentally different. Nano Banana, powered by Gemini 2.5 Flash, is optimized for high-speed image editing, inpainting, and style transfers, relying on iterative and conversational instructions. Its strength lies in rapid pattern-matching. In contrast, Nano Banana Pro, utilizing Gemini 3 Pro, is built for complex compositions, text rendering, and infographics, employing a structured and detailed prompting paradigm. Its advanced reasoning engine plans scene logic, and it can access real-time data via Google Search, offering high-fidelity output and superior text capabilities compared to Nano Banana's basic text rendering.

Nano Banana's semantic masking capability allows for intuitive image editing using natural language. Instead of manual selection, you can instruct the model to modify specific areas. * **Object Removal:** Use the 'remove' verb followed by a clear description of the element to be removed. To ensure other elements remain unchanged, explicitly state: 'Keep everything else in the image exactly the same, preserving the original style, lighting, and composition.' Example: 'Using this image, remove the stone bust from the foreground to create a clean, unobstructed view of the model and the garden. Keep everything else in the image exactly the same, preserving the original style, lighting, and composition.' * **Object Addition:** Define the new object and its placement. For complex scenes, describe its interaction with the environment, lighting, and perspective. Template: 'Using this image, add a [specific element] to the [location]. Ensure the new object matches the lighting and perspective of the original image.' Example: 'Using this image, add a regal Doberman Pinscher sitting obediently on the gravel path to the far left of the image. Ensure the new object matches the lighting and perspective of the original image.' * **Object Replacement:** This combines removal and addition in one step. Instruct the model to change only the specified element while preserving the rest of the image. Template: 'Using this image, replace the [old element] with a [new element]. Keep everything else in the image exactly the same, preserving the original style, lighting, and composition.' Example: 'Using this image, replace the stone bench with a sculptural, liquid-mercury-like flowing metal wave. Keep everything else in the image exactly the same, preserving the original style, lighting, and composition.'

Maintaining character consistency across multiple AI-generated images is challenging. The most reliable method with Nano Banana involves creating a 360-degree character sheet. This is a two-step process: 1. **Generate the reference sheet:** Create 2-3 images of your character from various angles (front, left, right, back) within a single frame or as separate generations. This provides the model with a comprehensive visual understanding of the character's features and attire. 2. **Use references for new scenes:** Once the character sheet is ready, use these images as references when prompting Nano Banana to place the character in diverse situations. This ensures stable proportions and consistent details. For example, to create a four-shot AI storyboard for a luxury watch ad, you would first generate a character sheet of a woman from different angles. Subsequently, you would use these reference images to prompt Nano Banana for various scenes, ensuring the character remains consistent. Example prompts for the storyboard might include placing the character in a luxury vehicle, on a villa terrace, watching a lake, or focusing on her hand and watch, all while maintaining hyper-realistic skin details and specific features.

Nano Banana Pro offers state-of-the-art text rendering, making it ideal for designs where typography is crucial. Unlike models that treat text as pixel textures, Nano Banana Pro uses a reasoning engine for high-fidelity, multilingual legibility. To achieve professional typographic results: * **Use double quotation marks:** Enclose the exact text string in quotes (e.g., “The Luxury of Being First”) to ensure literal rendering. * **Prioritize brevity:** Shorter phrases yield higher success rates. Limit text to a few words or sentences for optimal results. * **Guide the font style:** Use descriptive terms like 'clean, bold sans-serif' or 'elegant, traditional serif' rather than expecting exact font replication. * **Define the hierarchy:** Specify a 3-level text hierarchy: Level 1 Headline (large), Level 2 Subheaders (scanning-friendly), and Level 3 Body Copy (legible). Example prompt for a typographic infographic: 'Create a professional educational infographic titled “Choosing Your Typographic Tone.” Divide the layout into three horizontal sections, each representing a specific font personality: Serif (The Traditionalist): Use an elegant serif font to label this section. Include a brief caption: “Authoritative and elegant; perfect for formal or classic designs.” Add a small icon of a quill or a classic book. Sans-Serif (The Modernist): Use a clean, bold sans-serif font for the label. Include the caption: “Modern and tech-savvy; ideal for digital-first interfaces and clarity.” Add an icon of a smartphone or a geometric gear. Monospace (The Technicalist): Use a fixed-width monospace font for the label. Include the caption: “Analytical and functional; best for technical schematics or code-inspired looks.” Add an icon of a terminal prompt or a circuit board. Design Rules: Use a 3-level text hierarchy with a large Level 1 headline at the top. Allocate at least 30% of the canvas to white space to ensure a clean, professional look. Use a neutral color palette with bold accents for each section. Ensure all text is perfectly legible and correctly spelled.'

Nano Banana Pro integrates directly with Google Search, grounding its generations in real-world facts. This feature is invaluable for projects requiring accuracy, such as visualizing maps, historical events, or technical data. * **Factual accuracy:** Use the 'Thinking' model for visualizations that demand up-to-date information. * **Real-time context:** Prompt for current weather, specific landmarks, or verified historical details, and the model will confirm these facts before rendering. Example prompt: 'Use Google Search to check the current weather in Sydney, Australia, and add weather information on top of a cinematic landscape photo of the Sydney Opera House.' This live Google Search integration is a unique and powerful feature of Gemini 3 Pro, enhancing the reliability and relevance of AI-generated content.