You can conversationally tell Germini 25 Flash amage to edit only one part of an image while leaving the rest completely untouched.
Template
using the provided image, change only the specific element to (new elemene/deseniprion Keep everything else in the image exactly the same, preserving the original style lighong and composition
Example prompe
Using the provided image of a living rooms change only the blue sala to be a vintage, brown leather chesterloo Kop the rest of the room including the pillows on the sofa and the ghting unchanged.
Prompt
User uploads a logo (transparent or solid). First, analyze the logo's brightness and dominant color to select a background strategy: if the logo is dark/black/navy, set all backgrounds to Solid Matte White (Hex #F5F5F7); if light/white/silver/yellow, set all backgrounds to Deep Matte Charcoal (Hex #1A1A1A); if mid-tone/colorful, set all backgrounds to Soft Off-White (Hex #FOFOFO) for maximum readability. Use the same background for all four eras to maintain pitch deck consistency. The subject is a futuristic logistics hub on a floating island, styled as a low-poly 3D render with Claymorphism (Clay UI), soft occlusion shadows, Blender 3D Cycles rendering, and minimal texture. The logo should be integrated architecturally, not just as a billboard-for example, shaping the building, influencing road layouts, or appearing as a giant hologram. Generate four prompts for the following eras, always including the chosen background: The Industrial Dawn (1890s) with brick, steam, brass, and cobblestone; The Golden Age (1960s) featuring retro-futurism, curved shapes, and pastel accents; The Hyper-Present (2024) with glass, robotics, server racks, and LEDs; and The Solarpunk Future (2150) with organic white curves, trees, anti-gravity, and wind power
Provide an image and simply describe the change you want. The model will anotyze the original images style lighting and perspective to make the dit look nature and maintain character consistency across seves of images.
Template:
subject) please Joda/remove/modity to the is description of how the change should integrate Using the provided image of scerie Ensure the change
Example prompt
Using the provided image of my cat please add a small knitted wizard hat on its head Μαλά look like it's sitting comfortably and matches the soft lighting of the photo
11. "Same picture, but at night."
A man is standing in a modern electronic store analyzing a digital camera. He is wearing a watch. On the table in front of him are sunglasses, headphones on a stand, a shoe, a helmet and a sneaker, a white sneaker and a black sneaker
A cinematic 3/4 angle tracking shot follows a robot in the driver seat with both hands gripping a white steering wheel. The interior of the car is white with gold trim detailing. It's sleek and minimalistic. The robot is driving a retro futuristic sports car across a white sand desert canyon at noon.
Zoom Out
Multiple Angle Shots
ELS
LS
MLS
MS
MCU
CU
ECU
Low Angle
High Angle
Create a Nike ad with this image with text copy
Put him in a professional setting with a professional attire
1. "create a scene with this subject running in the existing environment, 16:9" 2. "create a scene with this subject entering into the building behind them, shot from behind, in the existing environment, 16:9" 3 "provide a 4-panel montage of b-roll footage of this subject, 16:9:
1. standing outside (back to the camera)
2. getting into the driver seat of a white sports car
3. getting into a matte gold horse-drawn chariot in the middle of the street
4. standing looking up towards the heavens with arms outstretched upwards (back to the camera)"
NANO BANANA PROMPT:
Change that nike sneakers into black and white Converse Chuck Taylor high-top sneaker
NANO BANANA PROMPT:
Replace the singer in the first image with the man from the second image. Keep the same dramatic pose
NANO BANANA PROMPT:
Create a Nike spec ad with 2 panel use the uploaded image as the main subject reference.
PROMPT:
Base Image > Shot 1
Prompt: Transform this scene into a peaceful morning atmosphere. Soft golden sunlight shines over the house
Shot 1 > Shot 2
Prompt: Transform this countryside house scene into a serene winter evening. Cover the ground, trees, and rooftops with snow, add light snowfall gently falling.
PROMPT:
Base Image > Shot 1
Prompt: Change the scene so the subject is standing at his home entrance, closing the wooden gate at night.
Shot 1 > Shot 2
Prompt: Change the angle to Aerial wide shot of the traditional Japanese home at night, softly lit with warm lanterns and surrounded by trees, no people in the scene.
PROMPT:
Base Image → Shot 1
Prompt: Change the lantern to shotgun
Shot 1 Shot 2
Prompt: the woman is no longer inside the barn.
Place her outdoors, sitting under a large tree in a forest clearing. She is holding the shotgun across her lap, carefully reloading shells into it with focused determination.
PROMPT:
Base Image → Shot 1
Prompt: Make them holding cocktail glass
Shot 1 Shot 2
Prompt: Change the girl in the middle become a man like tony stark ironman and they hug hum
PROMPT:
Base Image → Shot 1
Prompt: Make them sitting on billiard table
Shot 1 Shot 2
Prompt: Make them now playing billiard, 1 girl want shoot the ball, another one is standing waiting her turns
Prompt: Change the angle to a front view medium shot, with the whole body facing the camera.
Prompt: Make her running on the street, captured from a low-angle front view camera shot, dynamic motion blur on her body and hair, urban background, cinematic lighting, full body in frame, emphasizing power and energy.
Prompt: Change the angle to a front view medium shot, she is driving a car at high speed with a screaming expression, add motion blur effect, cinematic look.
Prompt: A fierce woman in a bright yellow Adidas tracksuit with white stripes and chunky sneakers, holding a katana raised above her head in a dramatic combat stance, inspired by samurai. Close-up medium shot, dynamic low angle, intense gaze partially obscure by the katan blade cinematic.
Prompt: Make her ride a motorcycle.
1. Bird's Eye View (Overhead)
View from directly above.
Gives you a flat-lay perspective, great for layout or full outfit reveal.
2. Worm's Eye View
Low angle looking upward.
Creates a feeling of dominance or power. Makes the subject feel larger-than-life.
3. Three-Quarter View
Slight side/front angle.
Adds depth and realism. Best for portraits and natural body posture.
4. Rear Three-Quarter View
Slight back/side angle.
Perfect for storytelling gives motion or attention to environment behind.
5. Over the Shoulder
Shot from behind, shoulder in view.
Great for POV-based designs or emotional narrative.
Reconstructing scenes with Nano Bananas
Using them in videos
Prompt: Generate a cinematic photorealistic sequence for my story. All frames are 16:9 aspect ratio
The Banana That Knew Too Much
Emi Kobayashi wasn't superstitious, but she believed in signs. When she found the banana behind the Inari shrine in Ueno Park, glowing faintly in the twilight, she knew it meant something.
"Don't touch me with your sweaty hands," the banana sald.
Emi screamed and fell backward onto the shrine steps.
"Oh, wonderful. Now you're going to have a bruise and blame me." The voice was female, sharp, and distinctly annoyed. "Could you not scream? I've had a very long day. Well, a very long four hundred years, technically."
Emi stared at the banana. It was brown in places, spotted with age, and wrapped in what looked like ancient rice paper bound with string. The characters written on the paper were faded but still visible.
"You're... a banana."
"Brilliant observation. And you're a teenage girl with a calculator watch and a Pokémon keychain. We both are what we are. Now pick me up before a crow gets ideas."
Despite every Instinct, Emi picked up the banana. It was warm, almost pulsing.
"Finally. I'm Chiyo. I've been waiting here since 1624 for someone from the Kobayashi family to come back. Do you have any idea how boring four hundred years is? I've memorized every prayer chanted at this shrine. Every. Single. One."
"How do you know I'm a Kobayashi?"
"Because you look exactly like the woman who cursed me, that's how. Same nose. Same Judgmental eyebrows. Are you going to help me or just stand there looking confused?"
Prompt: Generate a 9- image 'photo dump' grid of this person's vacation: a mirror selfie, a beach shot, friends at dinner, a blurry party photo, a scenery shot, a public bus shot, a pet moment, a sunset, and a candid smile
Recreate a new image from the point of you of the guy
Reimagine this shot but from Vader's point of view
"Take the clean modern living room as the base image. You're going to then replace the couch, coffee table, rug, and chairs with the furniture from the second beachy living room photo."
turn this into black-and-white manga"
↓
"make the next panel a funny cat-related twist"
This character is posing with the scythe raised high.
Add white lines to images
Let's add some comic book sound effects, like ZUBAAAAAN!!!!
"light animation of dog and cat gangsters."
"Flip stack of books to be upright and put on table between two bookends"
1. Show this photo as if someone is watching it through their living room TV.
2. make a lego version of this photo.
3. predict the next shot
4. A viral photo with some overlay like a YouTube thumbnail version which is basically saying "Nano Banana will break the internet" with an example right of this image and then what it can transform to in a nice way." but make sure to show right and actual transformation... show on the right hand side a version of me as a Simpsons character
@img1 I want to make this kind of YouTube thumbnail. Create four thumbnails for a Korean restaurant that grills Korean beef.
Make my face as if I was surprised. Also add my two hands on the sides of my face. Chage also the background to a blue gradient
Edit the uploaded thumbnail image while preserving its original composition, camera angle, framing, lighting direction, depth, proportions, and realism.
Strict rules:
-Do NOT change the pose, facial expression, head angle, or body proportions unless explicitly instructed.
-Do NOT introduce new objects or remove existing ones unless specified.
-Maintain photorealism and avoid Al artifacts, warping, or face drift.
Edits to apply:
Face replacement: Replace the original face with [YOUR FACE/ATTACHED PERSON], keeping the same expression, gaze direction, skin texture realism, and lighting match. Ensure perfect facial alignment and natural blending.
Outfit change: Replace the clothing with [DESCRIBE OUTFIT CLEARLY], keeping fabric folds, shadows, and body shape identical to the original.
Branding elements:
-Replace any existing logos with [YOUR BRAND LOGO/NAME]
- Replace all visible text with: "[YOUR HEADLINE TEXT]"
-Use [FONT STYLE: bold/condensed/modern/handwritten]
-Text color: [COLOR], with [stroke/glow/shadow] for readability
Color & mood: Adjust color grading to [cinematic/high-contrast / clean YouTube style/dark & dramatic] while preserving skin tones.
Background tweaks (optional): Replace or enhance background with [DESCRIBE BACKGROUND], keeping depth-of-field consistent with the original.
Quality requirements:
-Ultra-sharp thumbnail clarity
-No over-sharpening or cartoon look
-Perfect edge blending around hair and face
-Optimized for high-CTR YouTube thumbnail viewing on mobile
Output: A single, realistic, high-impact branded thumbnail that looks professionally
designed, not Al-generated.
It seems like the smoke was made into doodle parts...
change the lighting to night, and the headlights of the car are on
comic book strip style
Change the aspect ratio to use the aspect ratio from the blank image
Have all three women hanging around in a fun room. Use the aspect ratio from the blank image
Prompt:
Replace the car in the first image with the car from the second image, keeping the same characters, poses, and setting.
"Make it sunny and make her smile"
Ohh husbant meme
"Put the dog in the disaster meme"
Make the ramen bowl and ramen look like
2D whimsical hand drawn anime illustration keeping everything else in the image exactly the same
The woman in Figure 2 is wearing the necklace from Figure 1,Do not change the details of other Figure 2
Change the perspective to a high angle, looking down from above as if from the ceiling corner.
this is the north facade view of a house. generate the west facade view of this same house.
Change the camera perspective on your scene.
Front view → side view → behind view → overhead. Same subject, same moment, completely different shot.
Show me this scene from behind the subject.
Recreate this painting from the perspective of the barkeeper
Make this scene look like a lego
Prompt: "turn the manga to colour and translate to english"
Transform this drawing into a hyper-realistic 30 game render, ultra-detailed, cinematic lighting, dynamic composition, AAA video game style, realistic textures, depth of field.
Maintain the consistency and resemblance of up to 5 characters and the fidelity of up to 14 objects in a single workflow.
Switch between wide angle, close up, or panoramic shots. Change the depth of field to focus on different subjects.
Same scene, completely different shot.
Change color grades and lighting direction. Turn day into night. Switch moods instantly. Same scene, completely different feel.
Generate crisp visuals at 1k, 2k, or 4k resolution.
Switch aspect ratios for any platform - 16:9, 1:1, 9:16, whatever you need. Zoom in, resize, keep everything sharp.
Close-up shot on her face, eyes half-closed, subtle expression.
She is holding the phone visible but blurred.
Use the stick figure drawing as a reference guide. The stick figure shows the desired composition. Omit the stickman as an image inside the newly generated image.
Create pose sheet for this character: Standing front-facing holding bow, walking/striding side view in motion, running/jumping dynamic action pose, standing three-quarter view ready stance with bow, sitting on log resting pose with bow across lap
Transform this image to photograph style
Make this bird a mecha bird that is hellbent on causing my imminent harm
Animate this architectural sketch of Big Ben in London. Transform the static lines into a dynamic 3D scene where construction lines gradually fade, details sharpen, and the clock tower becomes realistic. Add subtle camera movement circling upward around the tower, trees swaying lightly, people walking in the square, and the clock hands ticking forward to bring the scene to life.
take the face of the person from the first image, dress them in the full outfit from the second image, and place them in the scene/background of the third image.
A hyper-realistic scene where SpongeBob, rendered as a believable 3D foam-like creature with soft porous texture and imperfect edges, is casually pushing a shopping cart in a real supermarket aisle under fluorescent lights, examining a box of cereal with exaggerated seriousness, blending seamlessly into the environment as if he physically exists among the shelves and shoppers.
Prompt : A photorealistic image, split down the middle. On the left side, a small, detailed bone that touches the center divider, resembling a dog bone set against a clear blue sky with clouds in background. On the right side, the bone seamlessly transitions into a detailed, futuristic stunning spacecraft, maintaining the line and shape of the bone against a deep space background scattered with stars. The spacecraft extends to the right edge, designed with elements that suggest a continuation of the bone's structure. The transition should be smooth, illustrating a creative blend where the organic shape of the bone becomes one with the mechanical structure of the spacecraft, set against a neutral background to emphasize the two subjects.
It can tum any sketch into an image perfectly with no fewn whatsoever
@img1 tsunami, dawn
@img1 post-apocalyptic
@img1 dawn,
sandstorm. No electricity
1st photo is original - "Expand this image by zooming out 1.5x and ensuring the aspect ratio is 16:9"
last prompt went for a higher view - "Expand this image by zooming out 1.5x and from higher angle and ensuring the aspect ratio is 16:9"
Take this location and make the landmark an isometric image (building only), in the style of the game Theme Park
Draw a DEM with contour lines" "draw the real world view from the red circle in the direction of the arrow
"draw what the red arrow sees"
prompt: place this man in the exact same position in a park soaking up the sun, realistic photo
If you're using Gemini without a text prompt, you'll need to select the image model; otherwise, it will sometimes start analyzing instead of generating.
This model has officially outperformed Seedream in realistic 4K images and is giving unbelievable 4K results.
Draw this cartoon image.
Make the process into six images in order!
A silent suburban yard under a moeniass sky.
The sliding beck door CREAKS open.
A MAN in his thirties steps out, half-awalie, pulling his robe tight against the cold. He pauses, listening
A faint HUM pulses somewhere in the dark.
He frowns, steps forward, and raises a FLASHLIGHT.
The beam cuts across the grass then stops.
A CRATER gepes in the middle of the backyard, fresh, smoking lightly.
The man edges closer, breath sharp, breath visible.
He sweeps the fleshlight downward.
At the center of the crater sits a glowing hall of light, hovering just inches above the torn earth. It radiates softly, shifting colors like a heartbeat.
The man tightens his grip on the flashlight.
MAN
(under his breath)
What the...
As he speaks, the glow begins to dim slowly, gently collapsing inward ke a dying star.
When the light fully fades, something remains:
A pretty girl, seated in the crater.
She wears a tight silver suit, metallic and smooth as liquid steel.
Her long white hair spills over her shoulders, shimmering faintly in the remaining smoke.
She sits perfectly still
The man just stares
Create a 3x3 grid of 9 photographs telling the story of a couple's wedding day, from preparation to celebration. Each photo should look like it was taken with a 1990's vintage film camera, with slightly faded colors, soft grain, and natural lighting. The grid should visually narrate the couple's journey, capturing emotions and moments in a cinematic, authentic way. The images should feel like real-life photojournalistic shots, not overly posed or artificial. Include a variety of angles, close-ups, and candid compositions. The nine photos should represent: morning preparations with the bride and groom getting ready separately, candidly adjusting clothes, hair, or accessories; a detail shot showing close-ups of wedding rings, bouquet, and other meaningful objects; the first glance with the couple seeing each other for the first time, capturing natural reactions and emotion; the ceremony with exchanging vows or rings, side angle showing expressions; the first kiss as a married couple, slightly wide shot capturing surroundings; walking together holding hands, candidly walking down the aisle or garden, joy visible; family & friends cheering, hugging, and celebrating, capturing candid emotions; reception moments with toasts, laughter, or dancing, showing natural joyful expressions; and a closing shot of the couple alone at sunset or under soft light, holding each other, intimate and emotional. All images should have consistent 1990's vintage film look: soft focus, warm faded tones, slight vignette, and natural grain, capturing emotions authentically with cinematic storytelling and aesthetic consistency across all 9 shots.
Use the Porsche car image as the exact subject reference. STEP 1 - VEHICLE RESTORATION & PREP: Before placing the car into the scene, fully restore it to pristine condition. Remove all dirt, dust, scratches, dents, damage, rust, and imperfections. Repair and perfectly align all body panels. Clean and polish the paint to a flawless, showroom-level finish. Restore headlights, taillights, glass, mirrors, rims, and tires to like-new condition. Maintain the car's original design, proportions, color, branding, and identity exactly. Do not modify the car's shape, model, or styling unless explicitly visible in the reference. STEP 2 - SUBJECT PLACEMENT & SCALE: Place the restored car naturally on a dirt forest path surrounded by wildflowers, using a low-angle perspective with a front three-quarter view. Position the car slightly off-center and dominant in the frame. The car should feel grounded and realistically integrated into the environment, with wheels lightly interacting with dirt and grass and no floating. STEP 3 ENVIRONMENT (EXACT SCENE MATCH): A serene alpine forest landscape during golden hour with tall evergreen pine trees surrounding the scene, a distant mountain range in the background with rocky textures and subtle snow patches, soft atmospheric depth separating foreground, midground, and background, a natural forest floor with dirt, grass, and small plants, and a foreground filled with out-of-focus wildflowers in pink, yellow, and purple creating depth. STEP 4 - LIGHTING & MOOD: Warm golden-hour sunlight coming from the upper right, soft rim light outlining the car's silhouette, natural sun bloom and gentle haze, balanced highlights on the car's curves with no blown-out reflections, cinematic contrast with soft shadows, and a calm, premium, aspirational mood. STEP 5 CAMERA & LENS: Professional automotive photography with a low camera angle, slightly wide lens at 35mm equivalent, shallow depth of field, sharp focus on the car, softly blurred foreground flowers, and background mountains slightly softened by atmospheric perspective. STEP 6 - MATERIALS & REFLECTIONS: Ultra-realistic paint reflections capturing trees, sky, and sunlight, clean and glossy yet realistic surface response, accurate glass reflections with subtle transparency, and tires detailed with realistic rubber texture. STEP 7 - COLOR GRADING & TONE: Natural, cinematic color grading with warm highlights, cooler greens in shadows, rich but realistic greens in trees, soft pastel wildflower colors in the foreground, and no oversaturation or artificial tones. STEP 8 - QUALITY & REALISM: Photorealistic, ultra-high resolution, film-quality realism with no CGI look, no plastic textures, natural imperfections only from real-world lighting, and professional automotive commercial quality. NEGATIVE PROMPT: cartoon, illustration, CGI look, toy car, distorted proportions, warped wheels, extra parts, incorrect badges, blurry subject, over-sharpening, oversaturated colors, unrealistic reflections, floating car, artificial lighting, studio background, city environment, people, text, watermark, logos added, low quality
Image Type: Surreal humorous character portrait. Subject: Pikachu - bright
yellow rodent, red cheeks, long ears with black tips, small arms. Hyper-realistic textures: soft fur, reflective eyes, subtle shading on cheeks. Head exaggeratedly oversized. Pose / Action: Standing in minimalist room, slightly leaning forward, cheeks puffed in comic surprise, small paws held near face. Setting: Simple interior with wooden floor, neutral walls, a small chair or table in background. Color Palette: Neutral room tones contrasting with vibrant yellow, red, and black
Lighting: Soft, natural lighting emphasizing fur texture and subtle glow of cheeks.
details. Emotion / Visual Storytelling. Cute, surreal exaghwetacreatesing energy and playful curiosity. Artistic Style: Hyper-realistic surreal digital art. Negative Prompts. Avoid overly cartoonish, blurry, or low-resolution textures.
Image Type: Surreal humorous character portrait. Subject: SpongeBob - yellow
porous sponge, white shirt, brown pants, red tie, large eyes. Hyper-realistic textures: sponge pores, subtle fabric wrinkles, reflective shoes. Oversized head exaggeration. Pose / Action: Standing slightly bent forward in a small modern room, hands resting on cheeks in a comedic overthinking pose. Setting: Minimalist interior with wooden floor, soft white walls, tiny table in the background. Lighting: Gentle indoor lighting emphasizing sponge texture and crisp clothing details. Color Palette: Neutral room tones with SpongeBob's vibrant yellows, brown pants, red tie, white shirt. Emotion/Visual Storytelling: Playful absurdity childlike wonder meets exaggerated surrealism. Artistic Style: Hyper-realistic surreal digital art. Negative Prompts: Avoid cartoonish flatness or messy, cluttered background.
Image Type: Surreal humorous character portrait. Subject: Popeye classic sailor with
exaggerated forearms, squinty eyes, pipe, white sailor hat, navy shirt, and brown pants. Hyper-realistic textures, rough skin, wrinkles, reflective pipe, subtle fabric folds. Oversized head exaggeration enhances comic absurdity. Pose / Action: Standing in minimalist interior, leaning forward slightly, one hand near chin as if pondering, the other resting at his side. Setting: Modern room with wooden floor, soft neutral walls, small table or chair in background. Lighting: Soft indoor lighting emphasizing skin texture, clothing folds, and pipe reflection. Color Palette: Neutral room tones with navy shirt, white hat, brown pants, skin tones, and pipe details. Emotion / Visual Storytelling: Comically thoughtful, playful exaggeration with hyper-realistic textures. Artistic Style: Hyper-realistic surreal digital art. Negative Prompts: Avoid messy background, distorted anatomy, or cartoonish flatness.
Image Type: Surreal humorous character portrait. Subject: Tom-gray-blue cat
with white muzzle, expressive green eyes, pink nose, and whiskers. Hyper-realistic fur, subtle shine on nose and eyes, oversized head for comic effect. Pose /Action: Standing in a modern interior, slightly hunched forward, one hand on chin in thoughtful pose, tail subtly curled. Setting: Minimalist room with wooden floor, small lamp or chair in background. Lighting: Soft indoor lighting highlighting fur texture and glossy eyes. Color Palette: Neutral room tones with gray-blue and white fur, pink nose accent. Emotion / Visual Storytelling: Overthinking, dramatic, playful surrealism - Tom in a comic existential moment. Artistic Style: Hyper-realistic surreal digital art. Negative Prompts: Avoid cartoonish flatness, distorted anatomy, messy background.
chat gpt
Ultra-cute 3D [baby deer] leaping toward the camera with outstretched paws, big glossy eyes, open smile, fluffy fur, tiny whiskers, red autumn maple leaves in background, golden-hour lighting, shallow depth of field, creamy bokeh, centered composition, cinematic colors, photoreal 3D render, ultra-sharp face focus, high detail
A Claymation-style [Subject], sculpted with visible fingerprints and subtle handmade imperfections. The character is [Action], with a joyful expression and soft stop-motion charm. Blurred mud background, props focus on the character in motion
Here are 5 subject/action pairs you can drop into your prompt:
Subject: a tiny astronaut
Action: bouncing in low gravity, arms outstretched mid-hop
Subject: a playful fox
Action: spinning in a twirl, tail swishing
Subject: a plump penguin
Action: shuffling forward with flappy little steps
Subject: a friendly wooden robot
Action: waving both hands while taking a clumsy step
0 Comments