Nano Banana Gemini Pro Guide: The Ultimate AI Image Workflow

Prompts from the Lecture

• "Can you remove the people who are standing on the beach?"

• "Make this in dramatic lighting and some of them are holding umbrellas."

• "A dramatic scene, professional shot, this man is standing next to this sport car."

• "Colorize the image."

• "Change the all clothes of yellow."

• "Turn this scene into a winter scene."

• "Make the woman sad."

More Effective Prompts for Nano Banana

1. Advanced Object Removal & Background Replacement

Instead of a general request like "remove the people," a more effective prompt specifies exactly what you want to remove and what should replace it, ensuring the final image is seamless and realistic.

Target Scenario: Removing people from a crowded beach and adding a new, serene atmosphere.

Prompt: Remove all people from the scene. Replace them with a tranquil, pristine beach landscape. Ensure the remaining elements like the water, sand, and sky are enhanced with a golden hour glow. Maintain consistent lighting and color harmony with the new, empty scene.

2. Detailed Scene Transformation with a Focus on Mood

This prompt goes beyond "dramatic lighting" by using specific terms to describe the desired mood and visual style, ensuring a more cohesive and professional result.

Target Scenario: Turning a sunny group photo into a moody, theatrical scene.

Prompt: Transform this scene into a noir-style photograph. Apply high-contrast, black-and-white lighting with deep shadows and sharp highlights. Add several people holding classic, full-sized black umbrellas to suggest a sudden downpour. Make the subjects' expressions more serious and contemplative to match the new somber atmosphere.

3. Sophisticated Composite Scene Creation

For combining objects, clarity is essential. This prompt breaks down the task into specific actions and provides visual cues to ensure the final image looks like a professional photoshoot.

Target Scenario: Placing a model next to a car in a high-end setting.

Prompt: Create a high-resolution composite image. Place the man from the provided photo standing confidently next to the sports car. Position him at a three-quarter angle, looking directly at the camera with a subtle smirk. Set the scene in a luxurious, modern parking garage at night. Use cinematic lighting with bright, targeted spotlights on the car and the man to create a sense of drama and opulence. Ensure the man's reflection is visible in the car's body panels for maximum realism.

4. Precise Object & Color Manipulation

Instead of a broad command like "Change the all clothes of yellow," a more effective prompt specifies which items to change and to what exact color, while also giving the AI instructions on how to handle the new colors.

Target Scenario: Uniformly changing the color of T-shirts in a group photo.

Prompt: Isolate all T-shirts in the image and change their color to a vibrant, warm saffron yellow (#FDB813). Ensure the new color is applied evenly, preserving the original fabric texture and shadows. Do not alter the color of any other clothing or objects.

5. Emotional and Environmental Transformation

This prompt demonstrates how to combine environmental and emotional changes for a more powerful and convincing transformation.

Target Scenario: Changing a happy summer selfie into a sad winter scene.

Prompt: Completely transform the scene into a stark, cold winter landscape. Replace the background with a snow-covered park and bare trees under a cloudy, gray sky. Adjust the subject's expression to one of quiet melancholy or sadness. Alter the lighting to be flat and muted, consistent with a winter day. Maintain the subject's pose and camera angle, but add subtle visual cues like a faint mist of breath.

Various Prompts for Different Scenes and Scenarios:

Now, let's explore a wider range of powerful prompts for your nano banana's editing abilities!

Scenario 1: Urban & Cityscapes

• Night to Day: "Transform this nighttime city skyline into a vibrant, bustling daytime scene under clear blue skies."

• Weather Change: "Overlay heavy, atmospheric rain onto this street scene, adding reflections on wet pavement and glistening surfaces."

• Futuristic Update: "Integrate futuristic elements into this city street: add flying vehicles, holographic advertisements, and sleek, modern architecture."

• Retro Aesthetic: "Apply a vintage, retro 1980s aesthetic to this urban landscape, with neon signs, classic cars, and a grainy film look."

• Crowd Dynamics: "Populate this empty city square with a diverse, energetic crowd of people engaged in various activities."

Scenario 2: Nature & Landscapes

• Seasonal Shift: "Change this autumn forest into a vibrant spring scene, with lush green foliage, blossoming flowers, and bright sunlight."

• Time of Day: "Convert this midday mountain view into a breathtaking sunset panorama, featuring warm golden light and elongated shadows."

• Atmospheric Mood: "Add a mystical, ethereal fog rolling through this ancient forest, enhancing the sense of mystery and depth."

• Wildlife Integration: "Place a majestic wild animal, such as a lion or a deer, naturally within this savanna/forest landscape."

• Environmental Restoration: "Remove all signs of pollution or human impact from this natural landscape, restoring it to a pristine, untouched state."

Scenario 3: Portraits & Figures

• Age Transformation: "Visually age this person by 20 years, showing realistic wrinkles and changes in hair texture." or "Visually de-age this person by 15 years, maintaining their identifiable features."

• Style Transformation: "Change the subject's outfit to a high-fashion, avant-garde design." or "Redress the subject in medieval fantasy armor."

• Hair & Makeup: "Give the person a completely new hairstyle: long, flowing, wavy red hair." or "Apply dramatic, artistic makeup to the subject's face."

• Background Swap: "Place the subject against a new background: a dimly lit, cozy library filled with books."

• Expressive Change: "Change the subject's expression from neutral to a joyous, wide smile with crinkling eyes."

Scenario 4: Objects & Still Life

• Material Alteration: "Transform the texture of this wooden table into polished, reflective marble."

• Object State: "Show this perfectly intact ceramic vase shattered into many pieces on the floor."

• Size & Scale: "Enlarge this small item (e.g., a coin) to an enormous, room-filling scale while maintaining realistic proportions."

• Color Palette Shift: "Recolor all elements of this still life composition into a monochrome blue palette, varying only in shades and tones."

• Environmental Interaction: "Show this object (e.g., a book) subtly levitating above its surface, casting a soft shadow."

Scenario 5: Artistic & Abstract

• Stylistic Render: "Render this photographic image in the style of a watercolor painting, with soft edges and blended colors."

• Geometric Abstraction: "Break down the elements of this scene into geometric shapes and vibrant, flat colors, inspired by Cubism."

• Dreamlike Quality: "Apply a soft-focus, hazy, and slightly distorted effect to create a dreamlike, surreal atmosphere."

• Glowing Elements: "Add a magical, internal glow to specific objects within the scene, making them appear luminous."

• Texture Overlay: "Overlay a subtle, intricate pattern or texture (e.g., cracked ice, flowing silk) across the entire image."

PROMPT LIST

• Original Prompt: "Apply a formal business outfit, modern office background and professional LinkedIn vibe."

  • Enhanced Prompt: "Generate a professional headshot for a LinkedIn profile. The subject should be wearing a well-fitted, modern suit with a neutral-colored shirt. The background should be a blurred, contemporary office space with soft, natural lighting. The overall vibe should be confident and approachable."

• Original Prompt: "Give me a confident Tinder photo with natural lighting and casual clothes."

  • Enhanced Prompt: "Create a confident and inviting social media profile picture. The subject should be wearing a stylish, casual outfit like a button-down shirt or a simple sweater. The setting should have soft, natural lighting, perhaps a bright cafe or an outdoor location. The expression should be genuine and friendly."

• Original Prompt: "Make this photo a stylish Facebook banner image, colorful and friendly."

  • Enhanced Prompt: "Transform this photo into a colorful and friendly Facebook banner. The background should feature vibrant, abstract patterns and cheerful, welcoming elements like flowers or subtle shapes. The text 'Hello World' should be integrated stylishly. Ensure the subject remains the central focus while the background adds a sense of warmth and creativity."

More Variation Prompts for Nano Banana

Here are additional prompts you can use to explore Nano Banana's character wizard capabilities, organized by category.

Professional & Business Prompts

• "Generate a professional photo for a corporate bio. Subject is in a smart casual outfit, seated at a desk with a laptop. Background is a clean, minimalist office with warm lighting. Focus on a knowledgeable and calm expression."

• "Create a headshot for an academic profile. Subject is wearing a tweed jacket, standing in front of a bookshelf. The style should be scholarly and serious, with warm, low-key lighting."

• "Produce a professional portrait for an artist or designer's portfolio. Subject should be in a creative, studio environment, wearing a stylish outfit with paint splatters or other subtle artistic elements. The lighting should be dramatic and artistic."

Social Media & Lifestyle Prompts

• "Generate a new Instagram profile picture with a vintage, retro look. Subject is in 70s-style clothing with long, wavy hair, standing in front of a colorful, sun-drenched backdrop."

• "Create a casual profile photo for a travel blog. Subject is wearing comfortable adventure clothes and a backpack, smiling while standing on a mountain trail with a sweeping view in the background. The lighting should be bright and outdoorsy."

• "Make a vibrant, pop art style profile picture. The subject's face should be clear, but the background should be filled with bright, cartoon-like colors and bold lines, similar to a Roy Lichtenstein painting."

Creative & Thematic Prompts

• "Transform this photo into a fantasy character portrait. Subject is dressed in medieval armor, with a sword on their back, standing in a misty forest. The lighting should be moody and mysterious."

• "Create a cyberpunk-themed character portrait. Subject is wearing a futuristic outfit with glowing accents, standing in a neon-lit, rainy street scene. The overall style should be dark and high-tech."

• "Produce a photo as if the subject were a character from a 1940s film noir. Subject is wearing a classic fedora and trench coat, standing in a dark alleyway with a single ray of light highlighting their face. The image should be in black and white with strong shadows."

PROMPT LIST: 

Step 1: All Original Prompts from Your Text

1. Turn this smoothie into an energy drink can featuring the elephant logo with wings.

2. Generate an influencer, she is holding this can in the studio professional lighting and shot.

3. Put this bottle in a luxury restaurant.

4. A woman is holding this bottle and she is wearing a black dress in this modern restaurant.

5. Close shot of one bottle and model is wearing a bathrobe.

6. (Whiskey case) A top-down, high-angle shot of a rustic oak barrel head… [full detailed scene with bottle, glasses, ice cubes, cigar].

Step 2: Super Prompts (Nano Banana Expert Edition)

Now let’s make these stronger, flexible, and reusable—like templates. I’ll enhance them with style, mood, and quality settings so anyone can use them across products.

Product Transformation (Smoothie → Energy Drink)

Prompt:
“Transform this smoothie photo into a branded energy drink can design. The can should feature a bold elephant logo with wings, sleek metallic finish, and vibrant product colors. Studio-style lighting, glossy reflections, and clean background suitable for advertisement.”

Adaptable for: soda, water bottles, coffee cans, etc.

Influencer Holding the Product

Prompt:
“Create a studio-shot of a modern influencer holding this product can with confidence. She is in a bright professional setting, with high-end studio lights creating glossy highlights on the product. The shot should look like a premium lifestyle advertisement, perfect for social media campaigns.”

Adaptable for: skincare, tech gadgets, luxury food products.

Luxury Restaurant Placement

Prompt:
“Place this product bottle elegantly on a luxury restaurant table. Surround it with fine dining details—polished cutlery, soft candlelight, blurred background with upscale décor. The atmosphere should feel premium, cinematic, and advertisement-ready.”

Adaptable for: wine, perfumes, designer packaging.

Lifestyle with a Model in Luxury Setting

Prompt:
“Generate a professional lifestyle photo: a woman in an elegant black dress is holding this product naturally in a modern luxury restaurant. The background should feature warm lighting, stylish furniture, and a sophisticated atmosphere. The image should feel like a luxury brand magazine ad.”

Adaptable for: jewelry, fashion accessories, luxury drinks.

Skincare Close-Up with Model

Prompt:
“Close-up product shot: one bottle in sharp focus, with a model wearing a soft white bathrobe in the background. Lighting is clean and bright, emphasizing freshness and self-care. Minimalist setting, glossy reflections on the bottle, lifestyle-oriented composition.”

Adaptable for: shampoos, lotions, cosmetics, wellness items.

Whiskey & Lifestyle Scene (Detailed Case)

Prompt:
“Top-down, high-angle professional studio shot of a rustic oak barrel head used as a serving tray. On the left, a full clear whiskey bottle with deep amber liquid. On the right, two short glasses filled with the same drink, ice cubes melting, golden reflections. A smooth cigar rests casually across one glass. Cinematic lighting, premium textures, styled for advertisement.”

Adaptable for: beer, wine, cocktails, cigars, luxury lifestyle branding.

These “super prompts” are flexible frameworks. You can swap product type, setting, or mood to fit different industries—skincare, fashion, food & beverage, luxury goods, tech, etc.

Original Prompts from Your Text

1. “Change the scene this girl is trying to open the door at night time.”

2. “Change the scene with closer shot and woman is looking at the camera.”

3. “This man is standing next to bar table in the same cafe.”

4. “This man is leaving from this cafe. Servant is showing him exit door.”

5. “Generate an image of this woman is leaning next to a beach bar and wearing casual clothes.”

6. “And add this guy into previous image. He is also wearing shorts and shirts and he is listening to music.”

7. “This man and woman are dancing next to the beach bar and there is a man playing a guitar with cowboy hat.”

Step 1 — Understanding the Goal

1. Your model or character photo (the base image), and

2. One of the outfit images you created earlier.

Step 2 — Generate 7 Core Outfit Styles

1. Bohemian Classic – flowy, natural fabrics, earthy colors.

2. Business Professional – sharp tailoring, modern confidence.

3. Casual Everyday – clean, relaxed, lifestyle wear.

4. Futuristic / Cyberpunk – bold metallics and neon tones.

5. Streetwear Urban – trendy, layered, youthful.

6. Luxury Evening – formal elegance and high-end fabrics.

7. Cultural Fusion – creative mix of modern and traditional.

Step 3 — Create Your Outfit Prompts

Bohemian Classic Outfit, you might use:

“A flat-lay or mannequin-style outfit layout showing a complete bohemian classic ensemble — a flowy patterned blouse, earthy-toned maxi skirt, leather sandals, woven crossbody bag, and layered jewelry with beads and stones. Include a wide-brim hat and round sunglasses.

Luxury Evening Outfit, here is the reference prompt.

“A glamorous luxury evening outfit reference — elegant dress or tailored tuxedo, high heels or polished shoes, clutch bag, jewelry, and matching accessories.

Arrange neatly on a dark velvet or marble background with soft golden light.

Focus on premium fabrics (silk, velvet, satin) and color harmony.

The photo should look like a high-end magazine editorial outfit preview

Step 4 — Upload Your Base Model

Make sure the photo has:

• Clear lighting,

• A neutral pose (preferably standing),

• Visible full body if you want a full outfit replacement.

Full Body Model Prompt

Take the uploaded character woman and create a full-body portrait of her standing naturally in a neutral pose. Capture her in a professional studio setting with soft, even lighting and a clean light-gray background.

Keep her appearance exactly the same — same face, hair, proportions, and overall identity. Just extend or render her as a full-body figure if necessary, ensuring the entire body is visible from head to shoes.

Make sure the lighting is balanced, shadows are soft, and the image looks high-quality and photorealistic.

8K resolution, DSLR studio realism, sharp details, true-to-life textures, and smooth natural shadows — ideal as a base image for applying different outfits later.

Step 5 — Combine the Model and Outfit (Combo Prompt)

You’ll upload two images into your AI tool:

• Image 1: Your base model,

• Image 2: The outfit image you just created.

Then use this combo prompt:

“Take the person from Image 1 (uploaded character) and dress them in the complete outfit from Image 2 (uploaded clothing reference). Keep the person’s pose, facial expression, hairstyle, skin tone, and body shape exactly the same as in Image 1. Integrate the outfit naturally and realistically — it should fit the person’s body perfectly, following natural body contours, fabric folds, and shadows. Ensure the lighting, shadow direction, and color tone of the new outfit match the environment and light source in Image 1 for seamless blending.”

Step 6 — Add-On Variations (Optional but Powerful)

• E-commerce / Catalog Look:
  Add a white background, sharp lighting, and emphasize textures for a clean product-style image.

• Street Style / OOTD Look:
  Place your model in a casual outdoor setting like a city street or café. Keep natural daylight and full-body framing.

• Editorial / Magazine Look:
  Go cinematic — adjust lighting, add atmosphere, and aim for elegance and depth.

• Costume / Themed Look:
  Replace clothing with historical, cultural, or fantasy costumes while keeping the person’s identity realistic.

Step 7 — Expand Your Outfit Library

You’ll find those in the “Attachment” section — both female and male outfit sets with full, detailed prompts.

Step 8 — Real-World Application Ideas

Here are a few ways you can apply this workflow:

• Create consistent fashion catalogs for online stores.

• Build AI character wardrobes for storytelling or brand identity.

• Design before-and-after outfit visuals for style guides.

• Experiment with thematic looks — from eco-fashion to luxury or futuristic.

How to Create Consistent Character Images with Nano Banana

Step 1: Create Your Base Model

We start by generating a base image — this is your foundation. It’s the version of your character that every future image will be built upon.

You can either:

• Use one of the ready-made prompts I’ll show you, or

• Upload your own photo to create a character that truly represents you or your brand.

Example for Woman:

“A stunning blonde woman with natural makeup, wearing a stylish modern casual outfit. Captured in soft daylight with a clean, aesthetic influencer style. Full-body composition with cinematic studio lighting, ultra-realistic skin texture, and sharp, detailed focus. Minimal white studio background, balanced tones, elegant posture, and modern lifestyle photography vibe — refined, photorealistic, and visually captivating.”

Example for Man:

“A handsome young man with a confident, natural expression and clean look, wearing a stylish modern casual outfit. Captured in soft daylight with a sleek influencer-style aesthetic. Full-body composition, cinematic lighting, ultra-realistic skin texture, and sharp detailed focus. Minimal white studio background, relaxed yet elegant posture — refined, photorealistic, and visually captivating.”

Step 2: Build Multi-Angle Reference Images

Here are the three prompts you’ll use:

1. Head Close-Up:

“Extreme close head shot of this same person.”
This helps the model learn the finer details of your character’s face.

1. Side Angle:

“Shot of the same person turning slightly to the left (or right).”
This gives the AI awareness of profile and side features.

1. Back Angle:

“Shot of the same person turning slightly from the back.”
This completes the 360-degree visual understanding.

Step 3: Generate Consistent Character Images in Any Situation

Now we can generate mode posses with various place and with various angles with our model.

Optimized Coffee Shop Prompt

Uploaded character is sitting at a sleek, modern wood coffee table, holding a white ceramic coffee cup with detailed latte art.

Candid influencer-style lifestyle photograph. Close-up to medium shot, eye level.

Warm, soft natural daylight streams in from a window, creating subtle rim lighting. Shallow depth of field (bokeh) with a soft, out-of-focus modern coffee shop interior background. Sharp focus on the character and the coffee cup.

High detail, hyperrealistic, 8k, professional photography, Canon EOS R5.

1. Coffee Shop Classic – “Urban Lifestyle Moment”

Prompt:
Uploaded character is sitting at a sleek, modern wood coffee table, holding a white ceramic coffee cup with detailed latte art. Candid influencer-style lifestyle photograph. Close-up to medium shot, eye level. Warm, soft natural daylight streams in from a window, creating subtle rim lighting. Shallow depth of field (bokeh) with a soft, out-of-focus modern coffee shop interior background. Sharp focus on the character and the coffee cup. High detail, hyperrealistic, 8k, professional photography, Canon EOS R5.

2. Luxury Wedding Scene – “Garden of Elegance”

Prompt:
Uploaded character is wearing a luxurious white wedding suit with fine tailored details, standing in an elegant open-air wedding garden filled with white roses, crystal chandeliers, and golden décor. The atmosphere is romantic and grand, with soft golden sunlight filtering through the trees and gentle breeze moving the floral arrangements. The scene is captured in cinematic daylight, shallow depth of field, with a bokeh background of guests and a white wedding arch. Hyperrealistic 8k photography, Canon EOS R5, premium editorial wedding style, perfect lighting balance, refined color grading, graceful and majestic mood.

3. Fashion Editorial – “Paris Fashion Week Studio”

Prompt:
Uploaded character is posing confidently in a luxurious Paris high-fashion studio, illuminated by expertly balanced professional lighting that highlights every detail with precision. The scene features a clean, minimalist set with soft gray backdrops and refined studio design. The character wears an avant-garde designer ensemble — sleek black and silver fabrics with impeccable tailoring, bold high-fashion accessories, and flawless runway makeup. Captured in a cinematic medium close-up with a shallow depth of field, every texture and contour appears crisp and luminous. The shot delivers a hyper-realistic, 8K Vogue-level editorial quality — photographed with a Hasselblad H6D for perfect color balance, tonal depth, and a sophisticated, high-end atmosphere.

4. Outdoor Travel Scene – “Old Italian Town Stroll”

Prompt:
Uploaded character is walking through a charming old Italian town with cobblestone streets, colorful historic buildings, and hanging flowers on balconies. They are wearing a smart-casual outfit — beige trousers, a white shirt, and a light scarf — carrying a small leather bag. The sunlight creates soft, warm reflections on stone walls and windows. The shot is candid, mid-motion, capturing authentic travel energy and lifestyle elegance. Cinematic travel photography, Leica M11, hyperrealistic 8k detail, warm vintage tones, relaxed and natural atmosphere.

5. Editorial Cinematic Scene – “Urban Night Lights” (my choice)

Prompt:
Uploaded character is standing under glowing neon lights in a modern city street at night. They wear a sleek black leather jacket, dark jeans, and subtle accessories, exuding a cinematic, mysterious energy. Reflections of red, blue, and purple lights dance on wet pavement, with passing cars creating beautiful motion blur in the background. Shot in ultra-realistic 8k detail, Sony A1, cinematic street photography style, shallow depth of field, rich color contrast, dramatic and stylish mood reminiscent of high-end movie scenes.

Creative Visual Story Infographic Prompt

Prompt:

Design a creative infographic that turns the following text into a visual story: [Insert your lecture or text here].
Combine text, icons, and simple illustrations to guide the viewer through the key ideas step-by-step.
Use color and imagery to create emotional engagement and visual rhythm — it should feel inspiring and easy to follow.
Add small design accents like arrows, connectors, or visual metaphors (e.g., mountain for growth, lightbulb for ideas) to strengthen understanding.
The tone should be friendly, motivational, and modern — similar to infographics shared by TED-Ed, Notion, or Canva’s educational templates.

Lecture:

“From Idea to Image: The Storytelling Formula for Gemini 2.5 Flash”

(Ideal for a creative, storytelling-style infographic)

The Concept

Think of Gemini 2.5 Flash as a visual storyteller. Your prompt is the story you tell it — the clearer and more emotional your story, the stronger the image.

1. Begin with the Scene

Start like a storyteller: Where and when does this happen?
Example: “Golden hour in a small coastal town — a painter captures the sea breeze on canvas.”

2. Describe the Character or Focus

Add your main subject: Who or what is central to this moment?
Example: “A young woman in a straw hat painting under the sunlight.”

3. Add Emotion or Mood

Emotion makes the image alive. Describe how it should feel — peaceful, dramatic, mysterious, joyful.
Example: “Peaceful, nostalgic atmosphere with warm light and soft shadows.”

4. Define the Visual Style

Mention your visual preference — realistic, 3D render, cinematic, watercolor, anime.
Example: “Cinematic film still, realistic texture, golden tones.”

5. Give Gemini a Creative Cue

End with a guiding hint, like:

• “Inspired by Pixar visuals”

• “Magazine cover composition”

• “Poster layout with central symmetry”

Final Touch: Each time you prompt, visualize the image before writing — Gemini mirrors your imagination almost exactly when you do this.

Infographic Title Suggestion:
? “The Storytelling Formula: How to Turn Your Imagination into Images with Gemini 2.5 Flash”

“Professional Educational Infographic Generator” Prompt

Create a professional infographic that visually summarizes the following text:

[Insert your lecture or text here].

Design it with a clean, modern layout suitable for educational or business presentation use.
Include key sections, main ideas, and supporting visuals (icons, charts, or diagrams) that make the content easy to understand at a glance.
Use a consistent color palette, balanced typography, and clear hierarchy to highlight the most important insights.
The style should feel clear, elegant, and visually engaging — like an infographic from Harvard Business Review or LinkedIn Learning.
Make sure the infographic has a logical flow (top to bottom or left to right) and strong visual balance.

? Lecture:

“Mastering Gemini 2.5 Flash Image Prompts: 5 Steps to Perfect Visuals”

(Ideal for a clean, structured educational infographic)

Overview

Gemini 2.5 Flash is Google’s advanced multimodal model capable of generating high-quality images from text prompts almost instantly. To get the best results, you need to communicate visually — not just verbally. Here’s how to do it effectively.

Step 1: Define the Core Scene

Start by describing what you want to see in one sentence. Keep it simple, like “a futuristic office with a team collaborating around holographic screens.” This gives Gemini a clear anchor for the image.

Step 2: Add Style and Mood

Next, describe how it should feel or look — for example, “cinematic lighting,” “modern tech aesthetic,” or “soft natural tones.”
Style words guide the artistic tone, helping Gemini match your creative intent.

Step 3: Set Perspective and Composition

Mention camera angles or framing: “wide shot,” “close-up portrait,” or “bird’s-eye view.” Gemini responds well to compositional cues that create structure and depth.

Step 4: Control Quality and Details

Use guiding phrases like “highly detailed,” “8K resolution,” or “realistic lighting.” Avoid overwhelming it with too many adjectives — 2–3 quality cues are enough.

Step 5: Refine and Iterate

If the first image isn’t perfect, adjust one variable at a time. Try changing the style, lighting, or emotion rather than rewriting everything. Gemini learns your visual taste through iteration.

Pro Tip: Add examples of artists, film types, or visual moods (like “Studio Ghibli watercolor” or “Blade Runner neon style”) to push Gemini toward a specific look.

Infographic Title Suggestion:
? “5 Steps to Master Prompting Gemini 2.5 Flash for Stunning Images”

Turn Complex Ideas into Educational Diagrams: Smart Educational Visuals

Prompt 1: The Water Cycle Explanation Diagram

Prompt:

Draw an educational diagram of the water cycle for academic presentation. Include realistic clouds, ocean, mountains, and arrows showing evaporation, condensation, precipitation, and collection. Add clear labels and short explanations near each stage. Keep the layout simple, colorful, and classroom-friendly.

? Tip: Try switching “water cycle” with carbon cycle or photosynthesis for instant variations.

⚡ Prompt 2: Electricity Flow in a Simple Circuit

Prompt:

Create a realistic educational diagram of a simple electric circuit with a battery, switch, bulb, and wires. Use arrows to show current flow direction and short annotations explaining each part’s function. The style should be clean, modern, and perfect for classroom slides or STEM materials.

? Pro Tip: Replace “simple electric circuit” with parallel circuit or solar-powered circuit for more advanced visuals.

Prompt 3: DNA Structure Diagram

Prompt:

Draw a DNA double helix diagram for educational presentation, with color-coded base pairs (A, T, G, C) labeled and explained. Include a short note showing how the strands twist and connect. Keep it realistic, detailed, and visually balanced with a clean white background.

? Variation: You can turn this into a chromosome structure or gene expression diagram by editing just a few words.

Prompt 4: Earth’s Layers Cross-Section

Prompt:

Design a cross-section diagram of Earth’s internal layers for academic presentation. Label the crust, mantle, outer core, and inner core. Use realistic textures and colors to show depth and temperature differences. Add neat annotations and arrows for clarity. Keep it clean, educational, and scientifically accurate.

? Try This: You can replace Earth with Mars or the Moon to create space science visuals instantly.

Prompt 5: Network Communication Diagram

Prompt:

Draw a computer network communication diagram for educational presentation. Include computers, a central server, routers, and arrows showing data flow. Label each component with short functional explanations. Use a sleek, professional style with a white or light-gray background, suitable for business or IT learning visuals.

? Pro Tip: Replace this with cloud storage system, AI workflow, or data security network for more specialized tech visuals.

Typographic Art with AI: Turning Words into Visual Design

1. Typographic Illustration Prompt – for creating visual objects or scenes from words.

2. Abstract Typographic Concept Prompt – for designing emotional, conceptual compositions using only letters.

Typographic Illustration Prompt

A Prompt Template:

Create a minimalist black-and-white typographic illustration of [describe the scene or object] using only the letters from the phrase “[insert phrase here]”.
Arrange and shape the letters to form the main elements of the scene and express motion or form.
No extra lines or shapes — only modified letters.
Keep it ultra-minimalist, balanced, and legible.

Example 1:

Create a minimalist black-and-white typographic illustration of a cat stretching on a windowsill using only the letters from the phrase “morning calm.”
Shape the letters to form the cat, the window, and a sense of quiet sunlight. Keep it simple, elegant, and entirely typographic.

Example 2:

Typographic Cityscape

Create a minimalist black-and-white typographic illustration of a modern city skyline using only the letters from the phrase “urban rhythm.” Arrange the letters to form buildings and streets with a sense of structure and motion — no extra shapes, crisp and legible.

Prompt 2: Abstract Typographic Concept

B Prompt Template:

Design a minimalist black-and-white abstract composition that visually represents [concept or emotion] using only the letters from the phrase “[insert phrase here]”.
Letters should be stretched, rotated, or arranged to capture the feeling or movement of the concept without losing legibility.
No additional shapes or graphics — only the modified letters.

Example:

Design a minimalist black-and-white abstract composition that represents the feeling of “serenity” using only the letters from the phrase “flowing silence.”
Arrange and distort the letters to express calmness and fluid balance — no extra lines, clean, and fully typographic.

Quick Overview

You’ll learn a clear 7-step process to go from AI prompts → images → storybook → social posting.

You’ll get:

• A set of six core prompts to generate your cinematic dreamer visuals.

• A workflow to refine, upscale, and adapt your images.

• A Gemini storybook-building prompt to create your visual narrative.

• A full process for social media posting and growth.

Step 1 — Define Your Visual Identity

The style:

• Black-and-white cinematic lighting

• Occasional color pop (like a yellow light, a red notebook, or blue neon sign)

• Soft emotional tones — introspection, solitude, inspiration

• Urban night scenes — cafés, alleys, windows, studios

Step 2 — Generate Images (Ready-to-Go Prompts)

Prompt 1 — The Midnight Walk

A cinematic black-and-white portrait of the uploaded character, a solitary artist walking through an empty street at midnight. The only vibrant color is the soft orange glow of a streetlamp reflected on wet pavement. The artist wears a long dark coat and carries a sketchbook under one arm. Their expression is thoughtful, eyes slightly turned toward the sky. The background shows soft bokeh lights of the city, blurred in drizzle. Moody cinematic lighting, introspective atmosphere, ultra-detailed 8K.

Prompt 2 — The Coffee of Creation

A cinematic black-and-white portrait of the uploaded character, seated in a small late-night café, surrounded by notebooks and half-empty coffee cups. The only vibrant color is the deep amber of the coffee in the cup. The character’s face is illuminated by a warm table lamp, eyes lost in thought. The window behind reflects city lights. Cinematic lighting, moody and poetic, ultra-detailed 8K

Prompt 3 — The Studio of Silence

A cinematic black-and-white portrait of the uploaded character, inside a quiet art studio. Canvas and brushes are scattered around. The only vibrant color is the soft blue glow of a computer screen or a desk lamp. The artist’s face is half in shadow, showing focus and fatigue. The background fades into soft darkness. Ultra-detailed 8K, cinematic lighting, introspective and still atmosphere.

Prompt 4 — The Window of Dreams

A cinematic black-and-white portrait of the uploaded character looking out from a tall window over the sleeping city. The only vibrant color is the golden reflection of the city lights in their eyes. The mood is poetic — quiet, reflective, full of longing. Soft cinematic lighting, moody depth, 8K resolution

Prompt 5 — The Spark of Inspiration

A cinematic black-and-white portrait of the uploaded character in a dimly lit alley, suddenly illuminated by a passing car’s headlights. The only vibrant color is the red notebook in their hand. The expression shows surprise and realization — as if a new idea was just born. Strong contrast lighting, cinematic drama, 8K resolution.

Prompt 6 — The Light Returns

A cinematic black-and-white portrait of the uploaded character standing under a glowing streetlight as dawn breaks. The only vibrant color is the faint pink of morning light on the horizon. The expression is calm, fulfilled, and inspired. Ultra-detailed 8K cinematic lighting, hopeful and serene atmosphere.

Step 3 — Iterate and Refine Images

Now that you have your six main images, let’s refine and expand them.

Generate 2–3 variations per prompt — explore angles, poses, lighting.

Then we’ll prepare your images for all aspect ratios:

• 1:1 for posts

• 16:9 for desktop views

• 9:16 for reels/stories

Aspect Ratio

Go to Google Whisk AI (from Google Labs). Upload your favorite images one at a time.
Then click the aspect ratio icon → choose your format (for example, 9:16 for Instagram Reels).

Upscaling

If you want higher resolution, use Canva’s Image Upscaler:

• Go to Canva → Create Design (16:9 or 1:1)

• Upload your image

• Use “Image Upscaler” under Apps → “Enhance”
  This improves texture and detail, making your visuals look professional.

Step 4 — Expand the Story with Variations

Variation 1 — At the Rooftop Edge

A cinematic black-and-white portrait of the uploaded character standing on a rooftop at dawn, the city lights fading below. The only vibrant color is a soft orange sunrise glow at the horizon. Calm yet powerful mood, ultra-detailed 8K.

Variation 2 — In the Library of Light

A cinematic black-and-white portrait of the uploaded character surrounded by stacks of books and soft desk lamps. The only vibrant color is the golden spine of one old book. Gentle, intellectual, inspiring tone.

Variation 3 — Under the Rain of Thoughts

A cinematic black-and-white portrait of the uploaded character standing under light rain, holding a closed umbrella. The only vibrant color is the reflection of a neon sign on wet pavement. Poetic, emotional, cinematic.

Step 5 — Create the Storybook (Gemini Workflow)

Open Gemini (or NotebookLM) and upload your 9 images in order.

Paste this prompt:

Storybook Creation Prompt

I will upload 9 cinematic black-and-white images.
Create a cinematic storybook titled “The Artist in Light and Shadow.”
Write 8 short chapters (each 70–120 words) inspired by the images.
Each chapter should feel poetic, emotional, and cinematic — about creativity, solitude, and inspiration.
The 9th image should serve as the cover or closing page with a reflective ending.
Maintain a consistent tone — introspective, film-like, visually rich.
Output only the storybook text, no layout notes.

Step 6 — Prepare for Social Media

Instagram Carousel:
1080×1350 (portrait) — use 5–6 images as a storytelling sequence.

Reels / TikTok:
9:16 ratio — use images + story text or a short voiceover.

What Google Whisk Can Do in 5 Minutes

1 What is Google Whisk?

Whisk is not just another text-to-image generator. It’s mostly an image-to-image tool, which means instead of writing long prompts, you lean on visual references. Think of it as telling the AI: “Here’s my style, here’s my subject, here’s my scene—make it happen.”

Under the hood, it combines three main Google AI models:

• Nano Banana: Focuses on editing and keeping your visuals consistent. Check “Precise Reference,” and Nano Banana makes sure your character or object stays the same across multiple scenes.

• Imagen: This is the engine generating the visuals themselves. Imagen 4 is your go-to if you want the highest quality.

• Veo/VO3: Want to animate your image? Veo handles the motion, letting you turn static visuals into short videos.

Access & Cost: Whisk is free with a Google account. Image generation is almost unlimited, but video clips are limited (usually 5–10 per month) unless you upgrade to AI Pro or Ultra.

2 Why Whisk is Powerful

Whisk isn’t just powerful—it’s smart and intuitive. Here’s why:

A. Visual Generation Through Context Images

Whisk organizes inputs into three categories, making image creation straightforward:

Subject - Scene - Style

Pro Tip: You can drag any generated image back into any slot for the next round. This is perfect for iterating fast.

B. Maintaining Consistency

One of Whisk’s biggest wins is consistency:

• Keep the same character in multiple scenes (e.g., your hero in Paris, on a beach, or in front of pyramids).

• If you want more creative freedom instead of strict consistency, just turn off “Precise Reference.”

C. Seamless Editing & Refinement

Forget downloading, tweaking, and re-uploading. With the Refine feature, you can type instructions directly:

• “Change the frog’s suit to a blue shirt.”

• “Add a briefcase next to the robot.”

You can also tweak your reference images before using them, like changing a color in a style image or adjusting textures.

D. Aspect Ratio & Output Control

• Landscape, portrait, square—you choose. Perfect for social media or presentations.

• Generate 2 images per run, up to 6 quickly with consecutive clicks.

• Use the dice roll button for random inspiration—it’s like brainstorming with AI.

E. Video Animation with Veo

Want motion? Click Animate, describe the movement, and Whisk makes a short video. Free users get limited clips per month, but it’s still a lot of fun for experiments or presentations.

F. Workflow & Sharing

• My Library: Everything you generate is saved automatically. No lost work.

• Recipes (Project Sharing): Share your exact inputs so others can remix or recreate your visuals.

3 Pro Tips for Getting the Best Results

1. Choose good references. The AI can only be as good as what you give it.

2. Use Subject slots wisely. Clear backgrounds help AI focus on your main object.

3. Leverage Nano Banana. Consistency is gold if you want your characters/products to stay recognizable.

4. Edit in-app. Use Refine rather than re-uploading for quick iterations.

Whisk Image & Animation Lesson (Quick Version)

1. Go to: labs.google → Click “Experiments” → “Create” → Select Whisk.

2. Upload 3 Layers:

   • Model (main subject)

   • Outfit & accessories (2nd subject)

   • Café background (scene)

3. Prompt:
   “Portrait of a woman sitting at a café shop.”
   Whisk merges all layers into one realistic photo.

4. Refine (Edit button )

   • Add: “...drinking her coffee.”

   • Or: “...wearing sunglasses.”
     Keeps the model consistent and natural.

5. Animate Mode

   • Click “Animate” → Enter: “She is drinking her coffee and says: what a lovely day.”

   • Whisk generates smooth, lip-synced animation with voice.

6. Style Experiment:

   • Upload animated-style image → Prompt: “This girl is singing and dancing.”

   • Try again without style → get realistic dancing version.

   • Animate it for a natural street video.

7. Bonus Tip:

   • Google G1 users get 5 free Whisk Animate videos/month.

   • Use them to experiment, practice, and create — free every month!

Goodbye Prompts. Hello Vision: Inside Google’s Whisk Revolution

What’s Google Whisk?

• Experimental AI tool by Google Labs (2025) — built for visual creators like designers, photographers, and illustrators.

• Mission: Create high-quality, consistent, and controllable AI visuals — without writing giant prompts.

• Instead of typing, you show the AI what you want using reference images.

Core Tech Power

• Imagen 4 → Creates the base image with realism and creativity.

• Nano Banana → Handles editing and consistency, keeping subjects, faces, or logos identical across images.

• Veo → Optional video generation engine (turns stills into motion clips).

Main Interface — The 3 Pillars

1. Subject: The main character or product.

   • Add multiple subjects, but  keeping them consistent together is still tough for AI.

2. Scene: The environment and lighting — e.g., futuristic studio, vintage cafe, medieval table.

3. Style: Defines artistic tone — from realism to cartoon to cinematic.

   • Example presets: Chocolate, Paper Cut, Sticker,  Photorealistic, 1960s Film Grain, Clay Render, Watercolor.

   • You can upload your own style image or use preset templates.

   • Only one active style (yellow checkmark) applies per generation.

The Yellow Checkmark

• Every image slot has a yellow checkmark .

• Only checked references are used in that generation — unchecked ones are ignored.

• Handy for testing combinations quickly without re-uploading.

Creative Shortcuts

• Dice Roll Button – Auto-generates a random idea or style if you’re stuck.

• Refine Mode – Chat directly with your image:

  • “Add sunlight from the left.”

  • “Make the shirt blue.”

  • “Add chocolate texture to the background.”

  • The image updates in place — no need to regenerate from scratch.

• Aspect Ratio Switcher: Instantly switch between 16:9, 9:16, and 1:1 for thumbnails, reels, or posts.

Prompt Rescue Hack

• When Whisk analyzes an image, it secretly writes a detailed internal description (hidden prompt).

• Click the little “info/edit” icon on your reference thumbnail → copy that description.

• Paste it into your main prompt box to make the AI fully respect that visual detail.

• Example: For a “1940s Film Noir” image, it might auto-write:

“High-contrast black-and-white lighting, deep shadows, visible film grain, soft vignetting.”
Paste that in to lock in the vibe.

Pro Workflow Trick

1. Generate multiple variations.

2. Pick the most accurate or best-looking result.

3. Drag that output back into the Subject slot and recheck it.

   • The AI now “learns” from its best version → improves accuracy and style consistency.

Animate with Veo

• Click “Animate” to turn stills into short AI videos.

• Add motion prompts like:

  • “Camera pans left.”

  • “The person smiles.”

• Free plan = ~100 credits/month (1 video ≈ 20 credits).

Organization & Sharing

• My Library / Projects Tabs: Save your subjects, scenes, styles, and prompts together.

• Shareable Recipes: Create and share “prompt setups” — your exact subject + scene + style combo in one click.

  • Great for teams or creative communities remixing each other’s ideas.

Limitations

• Image resolution currently capped around 1 megapixel — upscale for print work.

• Only one Subject, one Scene, one Style can actively guide the generation.

• Multi-character or complex group shots still inconsistent.

The Big Shift

• From writing long prompts → to curating visual references.

• The future of creative AI may look more like building mood boards than typing text.

• Best creators will be visual curators and iterators, not just prompt engineers.

IMPORTANT ANNOUNCEMENT: How to Access "Nano Banana Pro" for Free

All the prompts you will apply in this lecture is suitable for all Nano Banana versions.

Title: Unlock "Nano Banana Pro" & "Nano Banana 2.5" Directly on Gemini

Hello Class!

I have excellent news! You do not need any special developer software to access the advanced AI tools we discuss in class. You can access both the standard and the "Pro" versions directly on the regular Gemini website or app.

Here is the simple "Switch Trick" to get the best results for free.

1. The "Golden Rule" of Prompts

Please remember: All the prompts I provide in this lecture will work on ANY version.

• Nano Banana 2.5: Fast, reliable, and great for quick tests.

• Nano Banana Pro: Slower, but "thinks" deeper and gives higher detail.

• Result: You will never fail an assignment because you used the standard version. The prompt is the steering wheel, and it works on every car!

2. How to Access "Nano Banana Pro" (The Thinking Mode)

You can use the powerful "Pro" version for free with a daily limit. Here is exactly how to find it:

1. Go to gemini.google.com

2. Look at the top of the chat window for the Model Selector.

3. You will see two options (or a toggle):

   • Fast / Flash = This is Nano Banana 2.5. (Use this for unlimited, quick speed).

   • Thinking / Reasoning = This is Nano Banana Pro. (Use this for complex tasks).

The Secret: When you select "Thinking" mode, you are unlocking the Pro brain. And next from the left bottom from the tools select "Create Images". that is all. You can use Nano Banana Pro.

3. How to Generate "Pro" Images (The Banana Logo)

For our design lectures, the model you select changes the quality of your images.

• Step 1: Select "Thinking" mode at the top. Choose the create images from tools box.

• Step 2: Type your prompt (e.g., "Generate a cinematic 3D logo of a banana...").

• Step 3: Press Enter.

• Result: Gemini will use the Nano Banana Pro engine to create the image. It will look more realistic and follow instructions better than the standard version.

4. What are the Limits? (Free Version)

• Nano Banana Pro (Thinking Mode): You have a limited number of "Thinking" messages per day (usually around 5–10 for free users).

• What happens when I run out? Gemini will automatically switch you back to "Fast" (Nano Banana 2.5).

• Good News: You can continue working! The standard version is still powerful enough for all our exercises.

Summary:

• Start your day with Thinking Mode (Pro) for your most important prompts.

• Switch to Fast Mode (2.5) when you are just experimenting or if you hit your limit.

Let's create something amazing!

When you used Nano Banana before, it already felt powerful. You typed a prompt, and suddenly you had visuals that would normally take hours or days to create. That alone was impressive. But with Nano Banana Pro, something very important has changed—and once you notice it, you can’t unsee it.

The model doesn’t just generate images anymore. It thinks before it does anything.

And that small difference completely changes how you should work with it.

Think about how we’ve been using image tools until now. You tell them exactly what to do, and they obey. If the image looks strange, you fix the prompt. If the composition feels off, you try again. In other words, the intelligence lived mostly on your side. The model was talented, but passive.

Nano Banana Pro is not passive.

Now, before creating an image, the system tries to understand your intention. Not just what you typed—but why you typed it. Is this image for teaching? For branding? For persuasion? For storytelling? And based on that goal, it makes internal choices about layout, focus, realism, and even what details matter and what details should be ignored.

That’s what we mean when we say the image “thinks.”

It’s not about slowing things down or adding complexity. It’s about removing friction. You’ll notice that you need fewer corrections, fewer retries, and far less micromanagement. The image starts working with you instead of waiting for perfect instructions.

Another huge shift is that images can now explain themselves. This might sound simple, but it’s actually revolutionary. You can generate a visual and immediately understand why it looks the way it does. This is incredibly powerful for trainers, marketers, and leaders who need visuals that aren’t just beautiful, but defensible. It’s like having a designer who not only delivers the work, but also walks you through the design logic.

And then there’s awareness. Nano Banana Pro is no longer isolated from the world. It understands current styles, modern environments, real-world references, and up-to-date context. That matters more than people realize. A “modern office,” a “high-end brand,” or a “global team” means something very specific today—and now the model understands that without you having to over-explain.

You’ll also feel a massive difference in consistency. Characters remember who they are. Styles stay stable. Scenes feel like they belong to the same universe. This alone unlocks completely new possibilities—longer projects, brand systems, visual storytelling, even course materials that finally look coherent from start to finish.

And yes, the images are sharper. But what’s more important is that the detail now has purpose. Instead of random decorations, everything supports clarity. Text is easier to read. Visual hierarchy makes sense. Your viewer immediately knows where to look. That’s not just resolution—that’s intelligence.

So the biggest upgrade here isn’t a feature list. It’s a mindset shift.

Nano Banana Pro is no longer just an image generator. It’s a visual AI teammate. One that understands goals, context, and constraints. One that helps you solve problems visually instead of just making things look nice.

As we move into the next lectures, I want you to keep one simple idea in mind. Stop asking, “What image do I want?” and start asking, “What do I want this image to achieve?”

Nano Banana Pro is built for that second question—and once you learn how to speak its language, your entire workflow becomes faster, cleaner, and far more powerful.

Example Prompt (Multi-Frame Sequence – 4 Frames, Real Cinematic Style)

Prompt:
Create a cinematic image sequence of 4 frames.
A female journalist wearing a dark trench coat and carrying a notebook is walking through an old European city street during heavy rain at night.
Street lights reflected on wet cobblestones, realistic cinematic lighting, shallow depth of field, ultra-detailed, natural colors.
Aspect ratio 16:9.

Frame 1:
Wide establishing shot of the narrow street, rain pouring down, warm street lamps glowing, the journalist walking alone between historic buildings.

Frame 2:
Medium shot from the side as she walks, holding her notebook close, and raindrops visible in the air, light reflections on the pavement.

Frame 3:
Close-up of her face under the streetlight, rain on her hair and coat, focused and determined expression.

Frame 4:
Extreme close-up of her eyes, reflections of street lights and rain visible, cinematic depth and emotion.

Extreme Micro Cinematic Shot – Single Image Prompt

Prompt:
Create an extreme macro cinematic close-up of human eyes.
Ultra-detailed skin texture, visible eyelashes, fine wrinkles, natural pores.
Eyes sharply in focus, iris detail extremely high resolution, subtle reflections of soft street lights visible inside the eyes.
Shallow depth of field, cinematic lighting, realistic color grading.
Face partially visible, background completely blurred (bokeh), dramatic but natural mood.
Aspect ratio 16:9, ultra-realistic photography style.

What’s New: The Agentic Shift (June 2026)

Google has officially moved from the Chatbot Era (answering questions) to the Agentic Era (taking action on your behalf). The focus is now on proactive digital partners that plan, decide, and execute workflows independently.

• Gemini 3.5 Flash: The new flagship backbone engine built specifically for multi-step agentic workflows, complex long-horizon tasks, and lightning-fast code generation.

• Gemini Omni: A revolutionary native multimodal engine that can seamlessly process text, audio, images, and live video frames simultaneously to generate high-quality video or 3D environments.

• Gemini Spark: A 24/7 personal AI agent designed to autonomously run digital chores, manage operations, and anticipate daily user needs before being asked.

• NotebookLM Upgrades: Now natively running on Gemini 3.5, it features an isolated cloud computer to safely write and execute code, automates live web research to build source repositories, and directly exports full spreadsheets, data visualizations, and presentations.