Create and edit images with Gemini

Bring your imagination to life

Push design boundaries

Experiment with creative directions, or bring them into different contexts. Apply specific patterns to visible surfaces, or test out colors for fashion, design, and interior decoration.

Try in Gemini

AI-generated image of: Fashion portrait with a person in a voluminous red and white patterned garment and wide-brimmed hat, set against a solid royal blue background. The pattern appears to be stylized origami cranes, and the person wears blue lipstick.

Control the details

Create and edit images with powerful control. Replace the background, restore faded images, and change characters’ outfits. Keep tweaking until you’re happy, all with natural language.

Try in Gemini

AI-generated image of: A dramatic black and white photograph of a dancer silhouetted against bright backlighting, mid-leap. Large, flowing sheer fabric billows around the dancer, creating a circular, ethereal effect.

Generate, transform and edit images with simple text prompts, or combine multiple images to create something new. All in Gemini.

Capabilities
Performance
Safety
Try Gemini Image

Keep characters consistent

Reuse the same characters while changing their outfits, poses, the lighting, or the scene. Or reimagine yourself – across decades, in different places, or in your childhood dream job.

AI-generated image of: a side-by-side comparison demonstrating AI editing capabilities, showing an input image of a female astronaut wearing a helmet on the left, and the resulting image with the helmet removed on the right — Prompt: Remove the helmet

AI-generated image of: a layout demonstrating progressive AI edits. On the left is an input image of a woman with red hair in a car looking out at a desert landscape. On the right, a 2x2 grid shows cumulative edits: first removing the car's side mirror, then changing the landscape to snowy mountains, next dyeing her hair cool blond and magenta, and finally changing her green shirt to a yellow and blue flannel shirt. — Prompt 1: Remove the door mirror.
Prompt 2: Make the landscape snowy and mountainous.
Prompt 3: Make her hair dyed cool blond at the top and magenta at the bottom.
Prompt 4: She is wearing a yellow and dark blue flannel shirt

AI-generated image of: a side-by-side comparison. On the left is an input image of two simple blue cartoon characters. On the right, a detailed, vintage-style generated scene features these characters in a 1960s recording studio, with the larger one wearing headphones at a mixing console and the smaller one on a stool adjusting a reel-to-reel tape machine. — Prompt: A classic, faded photograph capturing a scene from a 1960s recording studio, featuring these two blue characters. They are depicted in the control room, surrounded by the warm glow of vacuum tubes and the complex array of a large-format mixing console. The larger of the two blue figures has a pair of bulky headphones placed slightly askew on its head and gazes peacefully through the soundproof glass at a musician in the live room. The smaller character, perched on a stool, wears a tiny pair of round, 1960s-style glasses and is turned slightly to adjust a knob on a reel-to-reel tape machine. The entire image has the aesthetic of an aged photograph, with a grainy texture, soft focus, and a desaturated, warm color palette.

AI-generated image of: a side-by-side comparison. On the left is a small input image of a woman wearing a delicate silver and pearl headpiece. On the right, the edited image shows the same woman wearing a large, elaborate headpiece made of vibrant red flowers, berries, and branches extending onto her face. — Prompt: Change head piece to something made from red flowers

AI-generated image of: a side-by-side comparison. On the left is an input portrait of a woman with curly hair. On the right, the generated output shows five Polaroid prints scattered on a light wood table, each featuring the same woman styled in various 1980s fashions and hairstyles. — Prompt: Create 5 headshot polaroid prints, laid out on a clean table, all of which show me in various situations from the 1980's

AI-generated image of: a side-by-side comparison. On the left is an input image of a woman wearing a vintage dress and bonnet sitting on an orange couch against floral wallpaper. On the right, the generated output depicts the same woman submerged underwater with floating hair and bubbles, completely replacing the original background. — Prompt: Make woman underwater, and remove the couch and wallpaper

Prompt: A close-up shot of a romantic moment holding each other while it snows

Prompt 1: Show this man as a teacher.

Prompt 2: Show this man as a sculptor.

Prompt 3: Show this man as a nurse.

Prompt 4: Show this man as a baker

AI-generated image of: a side-by-side comparison showing style transfer. On the left is an input photo of a smiling brown toy poodle. On the right, the dog is recreated as a large, stylized 16-bit pixel art character within a 2D platform video game level, complete with score, lives, and timer display. — Prompt: Recreate this dog as a 16-Bit Video Game character, and place the character in a level of a 2d 16-bit platform video game

AI-generated image of: a grid demonstrating variations of a cartoon T-rex. On the left is the input image of a smiling, stylized green T-rex toy on a black background. On the right, a 2x2 grid shows four variations of the T-rex wearing different costumes: a vampire with a cape and top hat, a superhero with a mask and cape, a fuzzy yellow chick costume, and a pirate with a feathered hat and costume. — Prompt 1: The t-rex is in a halloween costume.
Prompt 2: Now try a more fun costume.
Prompt 3: Fun. Now let's try a cute costume.
Prompt 4: How about a pirate costume?

Prompt, combine, create

Merge up to three images to create something new. Generate surrealist art, combine disparate photo elements, or seamlessly blend objects, colors, and textures.

AI-generated image of: a side-by-side comparison showing image merging. On the left are two separate input images: one of a swirling clear liquid against a blue background, and the other a lower-angle portrait of a woman in an orange plastic raincoat. On the right, the generated output shows a woman floating inside a massive, amorphous, translucent liquid-like bubble against a light blue background. — Prompt: A hyper-detailed, high-fashion photograph capturing a woman floating within a massive, amorphous bubble of translucent, glass-like liquid on light blue background.

AI-generated image of: a side-by-side comparison showing image merging. On the left are two separate input images: one of a snow-capped mountain at sunset, and the other of three humpback whales breaching in the ocean. On the right, the generated output merges the scenes, showing the whales breaching in the water with the large, snow-capped mountain as the backdrop against a sunset sky. — Prompt: Remix these 2 images

AI-generated image of: a side-by-side comparison showing image merging. On the left are two separate input images: one of a metal fork on a wooden table, and the other a bird's-eye view of a coiled circle of cooked spaghetti on a white plate. On the right, the generated output shows a single, oversized fork constructed entirely from tightly wrapped strands of spaghetti, resting on a white background. — Prompt: Remix these 2 images

AI-generated image of: a side-by-side comparison showing image merging. On the left are two separate input images: one of floating glass-like bubbles and the other of a couple sitting in a retro diner booth at sunset. On the right, the generated output merges the scenes, showing the couple in the diner surrounded by numerous large, reflective bubbles floating around them. — Prompt: Remix these 2 images

AI-generated image of: a side-by-side comparison showing image merging. On the left are two separate input images: one of two people in astronaut suits and the other a portrait of a woman looking out to sea with a surfboard nearby. On the right, the generated output merges the subjects, showing the man and woman lying down together in a field of colorful wildflowers, with the man wearing an open-faced astronaut helmet and the woman in a burgundy shirt. — Prompt: Replace the astronaut on the right with the women and remove the helmet on the astronaut on the left to show the man's face. The two are looking at each other.

AI-generated image of: a side-by-side comparison showing image merging. On the left are two separate input images: one of a glowing lightbulb and the other of a yellow banana. On the right, the generated output merges the two, showing a lit lightbulb base nestled inside a partially peeled banana, with the yellow peel curved to cradle the bulb against a soft yellow background. — Prompt: A Banana that peels to reveal a lightbulb

AI-generated image of: a side-by-side comparison showing image merging. On the left are two separate input images: one of a pink and white lotus flower floating on water, and the other a miniature view of synchronized swimmers forming a circle in a pool. On the right, the generated output combines the images, showing the tiny synchronized swimmers positioned inside the center of the large lotus flower, surrounded by water and lily pads. — Prompt: Combine these photos so that the synchronized swimmers are inside the lotus flower

AI-generated image of: a side-by-side comparison showing image merging. On the left are two separate input images: one of a man standing in front of a colorful graffiti wall, and the other of a German Shepherd dog sitting on a rocky path. On the right, the generated output shows the man sitting and hugging the German Shepherd, with a vibrant, abstract graffiti-style mural background. — Prompt: The man cuddles with his dog

Control the details

Create and edit images with powerful control. Replace the background, restore faded images, and change characters’ outfits. Keep tweaking until you’re happy, all with natural language.

AI-generated image of: a side-by-side comparison showing image editing. On the left is a small input image of a ballerina mid-leap, surrounded by swirling sheer fabric. On the right, the edited, full-size image shows the same ballerina in a different pose, with her arms dramatically raised above her head, still wrapped in the flowing fabric in a black and white image. — Prompt: Change the pose, ballerina is raising her arms

AI-generated image of: a side-by-side comparison showing image editing. On the left is a small input image of a decaying, abandoned grand hall with a grand piano. On the right, the edited, full-size image shows the hall completely restored to a pristine condition, with polished wooden floors and bright white architectural details, featuring the grand piano in the center. — Prompt: Make this environment completely brand new and clean, no decay

AI-generated image of: a side-by-side comparison showing image editing. On the left is a small input image of a woman in a yellow outfit standing in a minimalist architectural space defined by planes of bright red, yellow, and blue. On the right, the edited, full-size image shows the blue color changed to a bright green, creating a graphic composition of red, yellow, and two shades of green. — Prompt: Change all blue to green

AI-generated image of: a side-by-side comparison showing image editing. On the left is a small input image featuring a huge, dark tornado, with lightning and heavy clouds looming over a road, and a person and car in the foreground. On the right, the edited, full-size image shows the scene transformed to clear, bright weather at sunset with a vast blue sky, a rural road, and the person and car pulled over on the grassy shoulder. — Prompt: Remake with good weather

AI-generated image of: a grid demonstrating progressive image edits. On the left is an input image of a pink house in a suburban setting. The 2x2 grid on the right shows four stages of editing: first, the house is painted white; second, flower beds with vibrant blooms are added; third, the trees and landscape are changed to an autumn setting; and fourth, the scene is transformed into a winter setting with snow covering the ground and trees, and the house decorated for the holidays. — Prompt 1: The house painted white.
Prompt 2: Add flower beds with vibrant blooming flowers in front of the house.
Prompt 3: Transformed into a fall setting.
Prompt 4: Transform this image into a winter setting and decorate the houses

AI-generated image of: a side-by-side comparison showing image editing. On the left is a small input image of a blue and green bird taking off from water, wings spread. On the right, the edited, full-size image shows the same bird in the same action, but its primary body and wing feather colors are vividly changed to red with hints of emerald green. — Prompt: Recreate this bird as red with hints of emerald green

AI-generated image of: a side-by-side comparison showing image editing. On the left is a small input image of a golden retriever with its tongue out and head out of a moving car window. On the right, the edited, full-size image shows the dog with its mouth closed and a happy, open-mouthed smile, with its fur and ears blowing in the wind as it rides in the car. — Prompt: The dog’s mouth is closed

AI-generated image of: a side-by-side comparison showing image editing. On the left is a small input image of a smiling woman with bright red lipstick and orange earrings, partially obscured by a yellow curtain. On the right, the edited, full-size image shows a close-up of the woman, with the yellow curtain removed and replaced by a smooth yellow background that complements the orange earrings and red outfit. — Prompt: Remove the yellow curtain

Prompt: Restore photo

AI-generated image of: a side-by-side comparison showing image editing. On the left is a small input image of a city skyline at sunset from a rooftop. On the right, the edited, full-size image shows the scene transformed to daytime under a bright, clear sun, with the city skyline visible through a light haze. — Prompt: Change time of day to day with bright sun

AI-generated image of: a side-by-side comparison showing image editing. On the left is a small input image of a dilapidated, abandoned gas station with a rusty car and a faded red sign. On the right, the edited, full-size image shows the environment made vibrant and clean, with a clear blue sky, but maintaining the distressed look of the gas station structure and the rusty car, and correcting the sign to clearly read — Prompt: Fix the sign, make it say "GAS"

Push design boundaries

Experiment with creative directions, or bring them into different contexts. Apply specific patterns to visible surfaces, or test out colors for fashion, design, and interior decoration.

AI-generated image of: a side-by-side comparison. On the left is an input photo of a blue, black, and white butterfly on a red flower. On the right, the generated output shows a dark-skinned woman in a lavish, floor-length gown in the center of a lush botanical garden, with the gown's skirt patterned after the butterfly's wings in bright blue and black. — Prompt: Turn this into a stunning outfit of a woman walking through a luscious botanical garden

AI-generated image of: a side-by-side comparison showing a room restyling. On the left are an input image of a modern living room and a color palette swatch of gold, gray, purple, teal, and blue. On the right, the generated output shows the living room restyled in a fresh, dreamy style with light colors and textures, incorporating elements like a textured white sofa, woven pillows, and a rattan light fixture. — Prompt: Restyle this living room in a fresh dreamy style mixing woven materials and textures using the colour swatches

AI-generated image of: a side-by-side comparison. On the left is an input portrait of a smiling man. On the right, the generated output shows a fictional 1960s-style cereal box named — Prompt: Turn me into a cartoon like character on the front of a 1960's cereal packet of 'Adventure O's' along with other text you would find on a cereal box from the 1960's. the packet sits in a breakfast table in a photo reminiscent of the 1970's

AI-generated image of: a side-by-side comparison. On the left is an input image of architectural blueprints for an igloo. On the right, the generated output shows a completed, realistic-looking igloo built from snow blocks, sitting in a snow-covered arctic landscape under a clear blue sky. — Prompt: Show me what you can build with this blueprint

AI-generated image of: a side-by-side comparison showing a room restyling. On the left is an input image of a modern, simple bedroom. On the right, the generated output shows the bedroom completely transformed into a — Prompt: A bedroom restyled in an over the top, maximalism style of 'The Cassette-Futurist Room': A 1980s vision of the future. Think clunky, beige plastic computers built into the walls, furniture with unnecessary vents and chunky buttons, and a colour palette of grey, orange, and brown. It's the tech aesthetic seen in classic sci-fi films of that era.

AI-generated image of: a side-by-side comparison showing image merging. On the left are two input images: one of a red and white geometric pattern and the other of a mannequin wearing a yellow dress with blue speckles and a large sun hat. On the right, the generated output shows a close-up of a woman in a dramatic, high-fashion, red and white origami paper-patterned dress and large matching sun hat, against a bright blue background. — Prompt: Fashion shot of a woman in a huge dress in origami paper red and white geometric pattern style

AI-generated image of: a side-by-side comparison showing image variation generation. On the left is an input image of a postage stamp with a mushroom illustration. On the right, a 2x2 grid shows four different generated postage stamps, each featuring a unique red and orange mushroom illustration in the same stylistic dark background and perforated border as the input. — Prompt: Create another post stamp in the same series with a different mushroom

AI-generated image of: a side-by-side comparison. On the left is an input image of a red bird with ruffled feathers taking off from water. On the right, the generated output shows a woman in a dramatic, floor-length, feathered red gown standing in shallow water, with the dress's sleeves forming large wings, set against a background of a glacier lake and snow-capped mountains. — Prompt: Make this feathered texture into a stunning dress on a woman standing by a glacier lake

AI-generated image of: a side-by-side comparison showing image merging. On the left are three input images: a red and white leaf pattern, a blue and white geometric pattern, and a woman in a plain pink dress. On the right, the generated output shows the woman standing in a room with the red and white leaf pattern covering the walls and floor, while she is wearing a long dress with the blue and white geometric pattern. — Prompt: Woman walks in the dress with pattern from image 2 in the room with walls and floor from image 1

One prompt, many possibilities

Generate multiple images using just one prompt to explore different creative avenues. Or create several images that work together to tell a complete story.

An 8-part visual story, arranged in a grid, of two blue, round, cartoon-like characters (one large, one small) and their adventures in the 1960s music scene. **Image 1:** The duo arrives at a bustling city street, excited. **Image 2:** They perform their first gig on a stage with a band. **Image 3:** They achieve success, performing triumphantly on a larger stage. **Image 4:** A moment of sadness in a dressing room, the smaller character offering comfort. **Image 5:** The large character records music alone in a studio booth, the small one watches. **Image 6:** The large character performs solo in a small, dim club. **Image 7:** The large character is alone at night, writing music at a desk. **Image 8 (Finale):** The two characters reunite, performing triumphantly on stage under a — Prompt: Create a beautifully entertaining 8 part story with 8 images with two blue characters and their adventures in the 1960s music scene. The story is thrilling throughout with emotional highs and lows and ending on a great twist and high note. Do not include any words or text on the images but tell the story purely through the imagery itself.

**Input Image:** A black and white photo of a man and a woman with curly red hair, posing together; the man wears a dark jacket, the woman a cream sweater. **Image Grid:** A 12-part black and white film noir detective story. Image 1: Female detective points at a map on a desk. Image 2: Female detective studies a map. Image 3: The duo looks at something on a table. Image 4: They examine a large book in a library. Image 5: They stand in a dark alley. Image 6: They run down a wet city street. Image 7: They look out a window, the man distressed. Image 8: They examine a large vault door. Image 9: The woman operates the vault dial. Image 10: The open vault reveals treasure; they are surprised. Image 11: They stand outside a grand building. Image 12: They toast each other across a desk of case files. — Prompt: Create an addictively intriguing 12 part story with 12 images with these two characters in a classic black and white film noir detective story. Make it about missing treasure that they get clues for throughout and then finally discover. The story is thrilling throughout with emotional highs and lows and ending on a great twist and high note. Do not include any words or text on the images but tell the story purely through the imagery itself.

**Input Images:** Two separate headshots of the female superhero protagonists: one with dark braided hair and one with long black hair. **Image Grid:** A 9-part comic book-style story of two female superheroes fighting an enormous monster over a cityscape. Image 1: Protagonist with braided hair holding a glowing blue shield display. Image 2: Protagonist with long black hair holding a glowing blue shield display. Image 3: Both protagonists standing side-by-side in their green and blue superhero suits. Image 4: The two heroes face a giant, shadowy, monstrous creature with green energy. Image 5: The hero with black hair flies towards the monster, shooting a blue energy beam. Image 6: The monster is struck by green lightning-like energy over the city. Image 7: A glowing green dome or sphere of energy erupts over the city. Image 8: The two heroes fight inside a large, glowing green-blue energy sphere, a decisive moment. Image 9 (Finale): The two heroes stand victoriously together on a rooftop overlooking the cityscape, the monster defeated. — Prompt: Create a riveting epic 9 part story with 9 images with these two protagonists and their adventures as secret super heros. The story is thrilling throughout with emotional highs and lows and ending on a great twist and high note. Do not include any words or text on the images but tell the story purely through the imagery itself.

Capabilities

prompt_spark

Multimodal understanding

Upload images and share text instructions with Gemini to create complex and detailed images.

chat_spark

Conversational inputs

Use everyday language while creating images, and keep the conversation going to refine what the model generates.

globe

Real-world knowledge

Generate images that follow real-world logic, thanks to Gemini’s advanced reasoning capabilities.

Performance

Gemini 2.5 Flash Image is a state-of-the-art image generation and editing model, with lower latency compared to other leading models.

Gemini 2.5 Flash Image was tested on LMArena as nano-banana.

chevron_right

View model card

Scatter plot titled 'Text to Image: Overall Preference on LMArena' comparing models by Elo score (y-axis, higher is better) and throughput in pixels per second (x-axis). Gemini 2.5 Flash Image is the top performer with the highest Elo and strong throughput. Gemini 2.0 Flash Image has the highest throughput but a lower Elo. ChatGPT 4o / GPT Image 1 (High) has a high Elo but the lowest throughput. Other models shown include Imagen 4 Ultra 06-06 and FLUX.1 Context [max].

Two charts titled 'Text to Image'. The left chart shows Elo scores with 95% confidence intervals for 'Overall Preference', 'Visual Quality', and 'Prompt Following'. Gemini 2.5 Flash Image has the highest Elo in Overall Preference. The right chart shows 'Percentage of substring matches' for 'Text Rendering', where Gemini 2.0 Flash Image scores highest (72.7%), followed closely by ChatGPT 4o (71.8%). Other models compared include Imagen 3 Ultra, FLUX.1, and Gemini 2.0 Flash.

Bar chart titled 'Image Editing' comparing Elo scores (with 95% confidence intervals) across seven categories: Overall Preference, Character, Creative, Infographics, Object Environment, Product Recontextualization, and Stylization. The models compared are Gemini 2.5 Flash Image, ChatGPT 4o / GPT Image 1 (High), FLUX.1 Context [max], Qwen Image Edit, and Gemini 2.0 Flash Image. Gemini 2.5 Flash Image generally records the highest scores across most categories, notably in Overall Preference.

Limitations

While Gemini can now create a wide range of images, we’re still working on improving key capabilities.

Factual representation

Not every image Gemini generates will be perfect – it can still struggle with small faces, accurate spelling, and fine details in images.

Character features

The model excels at character consistency, but it may not always get it right. We're working to make this consistency even more reliable.

Safety

We use extensive filtering and data labeling to minimize harmful content in datasets and reduce the likelihood of harmful outputs. We also conduct red teaming and evaluations on content safety, including child safety, and representation.

Image generation in Gemini has all our latest privacy and safety features. This includes SynthID, our tool that embeds an invisible digital watermark directly into an image, allowing it to be identified as AI generated.

chevron_right

Learn more