Google just shipped Nano Banana 2 — the image model behind Gemini’s photo editing — and it changes what’s possible with a text prompt. No layers. No export settings. No YouTube tutorial about masking. You describe what you want, and the AI rebuilds your photo.
This isn’t a gimmick. Nano Banana 2 combines the studio-grade quality of the Pro model with Flash-level speed. It understands lighting physics, skin texture, color theory, and spatial depth — all from a sentence you type into the Gemini app.
Below are 30 prompts organized into six categories. Bookmark this page. Use it as a reference sheet. Copy the prompts exactly, or remix them into your own style — the best results come from stacking and combining.
How to Use These Prompts
Open the Gemini app. Upload your photo. Paste any prompt below into the chat. Nano Banana 2 handles the rest — typically in under 10 seconds.
Three rules that separate good results from great ones:
Be specific about what to keep. Telling the AI what NOT to change matters as much as telling it what to change. “Maintain natural skin tones” prevents your portrait from looking like it was dipped in orange paint.
Stack prompts for custom styles. Run a lighting fix first, then apply a cinematic grade second. Layering simple prompts beats writing one massive paragraph.
Name the vibe, not just the technique. “Golden hour glow” gives the AI more context than “make it warmer.” Nano Banana 2 has been trained on millions of photographic styles — reference them.
Category 1: Fix Lighting, Blur, and Quality
These five prompts rescue photos that most people would delete. Bad lighting, camera shake, grainy darkness — fixable in seconds. Start here before applying any creative edits.
1. Fix Exposure and Hit DSLR Quality
When to use it: Your photo is too dark, too bright, or just looks like it came from a 2015 Android phone.
Enhance this photo by correcting exposure, balancing highlights and shadows, and improving overall brightness while maintaining natural color tones. Sharpen fine details, reduce noise, and improve clarity across the entire image. Ensure realistic textures and skin tones, and output a high-resolution, DSLR-quality result with a clean, professional finish.
Why it works: The prompt addresses five problems simultaneously — exposure, noise, sharpness, color, and texture — while anchoring everything to “natural” and “realistic.” That anchor stops the AI from overcooking the edit.
2. Rescue Low-Light Photos
When to use it: Indoor shots, evening photos, or anything taken without flash that came out muddy and grainy.
Fix low-light conditions by intelligently increasing brightness without overexposing highlights. Reduce grain and digital noise, enhance subject visibility, and apply soft cinematic lighting that feels natural. Balance color temperature, refine shadows, and create a clear, visually appealing image with professional-level quality and smooth tonal transitions.
Why it works: “Without overexposing highlights” is the key phrase. Most brightness fixes blow out the whites. This prompt tells Nano Banana 2 to lift the darks while protecting the brights — exactly what a professional colorist does manually.
3. Sharpen Clarity Across the Frame
When to use it: The photo is fine but feels soft. Details lack punch. Everything looks slightly out of focus.
Improve overall image clarity by sharpening edges, enhancing textures, and refining fine details. Adjust contrast slightly to create depth while maintaining natural tones. Ensure the subject remains crisp and well-defined, and produce a polished, high-quality output that resembles professional camera photography without looking over-processed.
Why it works: “Without looking over-processed” prevents the HDR nightmare where every blade of grass glows. The contrast-for-depth instruction adds dimension without touching saturation.
4. Turn Blurry Into Sharp HD
When to use it: Camera shake, subject motion, or autofocus failure ruined an otherwise good shot.
Convert this blurry image into a sharp, high-definition version by restoring lost details and refining edges carefully. Avoid artificial sharpening artifacts, maintain natural textures, and improve overall clarity. Enhance lighting and contrast subtly to support detail recovery, creating a realistic and professional-looking final image.
Why it works: The “avoid artificial sharpening artifacts” instruction matters. Without it, Nano Banana 2 might add halos around edges — the telltale sign of cheap sharpening. This prompt prioritizes reconstruction over brute-force enhancement.
5. Balance Highlights and Shadows
When to use it: Harsh sunlight, backlit subjects, or any scene where parts are too bright and other parts are too dark.
Enhance dynamic range by recovering details in both shadows and highlights while maintaining natural balance. Adjust mid-tones for clarity, refine contrast for depth, and ensure smooth tonal transitions. Preserve realistic colors and textures, resulting in a visually balanced, high-quality image that feels natural yet professionally edited.
Why it works: This is the AI equivalent of pulling the highlights slider down and the shadows slider up in Lightroom — but with intelligence about what “natural balance” means for the specific scene in your photo.
Category 2: Cinematic Color and Mood
These prompts turn ordinary photos into something that looks like it belongs in a film. Color grading is the single biggest difference between amateur and professional-looking images — and it’s where Nano Banana 2 genuinely shines.
6. Apply a Cinematic Movie Look
When to use it: You want your photo to feel like a still frame pulled from a film. Moody, dramatic, intentional.
Apply cinematic color grading with moody tones, soft shadows, and subtle contrast. Introduce a film-like atmosphere with gentle grain, balanced highlights, and controlled blacks. Ensure skin tones remain natural while creating a visually dramatic, storytelling-driven image that feels like a frame from a high-quality movie.
Why it works: “Controlled blacks” is doing heavy lifting here. In cinema, blacks are rarely pure black — they’re lifted slightly, giving that signature film look. Combined with “gentle grain,” this produces the organic texture that digital photos lack.
7. Create a Dramatic Scene
When to use it: Portraits, street photography, or any shot where you want intensity and visual weight.
Transform this image into a dramatic movie-style scene by adding directional lighting, deep shadows, and strong contrast. Emphasize subject focus while creating depth and mood. Use color grading that enhances storytelling, making the image feel intense, cinematic, and visually engaging like a professional film still.
Why it works: “Directional lighting” gives the AI permission to reshape where light falls in your image. Combined with “deep shadows,” this creates the three-dimensional pop that flat phone photos never have.
8. Hollywood Teal-and-Orange Grade
When to use it: You want that blockbuster color palette — warm skin against cool backgrounds — that dominates modern cinema.
Apply teal and orange cinematic color grading with careful attention to skin tones and natural balance. Increase contrast slightly, enhance depth, and refine highlights for a polished look. Ensure the image feels vibrant yet controlled, delivering a high-end Hollywood-style finish with strong visual appeal.
Why it works: Teal and orange is the most widely used color grading combo in Hollywood because warm skin tones pop against cool blue-green backgrounds. Naming the exact color scheme tells Nano Banana 2 exactly which look to aim for.
9. Dark Moody Aesthetic
When to use it: Instagram feeds with a dark, cohesive visual identity. Fashion, lifestyle, or product photos that need atmosphere.
Create a dark aesthetic mood by lowering brightness slightly, deepening shadows, and adding soft glow highlights. Use a minimal color palette with subtle tones to maintain consistency. Ensure the subject remains visible and sharp while achieving a modern, Instagram-friendly aesthetic that feels artistic and premium.
Why it works: “Minimal color palette” forces the AI to desaturate selectively rather than just slapping a dark filter over everything. The “subject remains visible and sharp” instruction prevents the edit from burying your subject in shadow.
10. Instant Golden Hour
When to use it: You missed the magic hour, or the lighting was flat, and you want that warm, sun-drenched glow.
Add warm golden hour lighting with soft sunlight glow and gentle highlights. Adjust color temperature toward warm tones, enhance natural light reflections, and create a dreamy outdoor atmosphere. Maintain realistic skin tones and balanced exposure while delivering a visually pleasing, soft, and cinematic final image.
Why it works: “Enhance natural light reflections” tells the AI to work with existing light sources in the image rather than painting warmth over everything uniformly. The result looks like the sun cooperated instead of like you applied an Instagram filter.
Category 3: Portrait and Beauty
Portrait editing is where AI tools used to fail hardest — plastic skin, uncanny eyes, vanishing pores. Nano Banana 2 is dramatically better at preserving facial identity while enhancing everything around it. These prompts are calibrated for that balance.
11. Smooth Skin Without the Plastic Look
When to use it: Portraits where you want cleaner skin but still want the person to look like a person, not a mannequin.
Enhance this portrait by smoothing skin naturally while preserving texture and avoiding an artificial look. Sharpen eyes, refine facial details, and improve lighting balance. Add soft studio-style lighting and subtle contrast to create a professional, polished portrait with high clarity and natural beauty.
Why it works: “Preserving texture” is the critical instruction. Without it, AI skin smoothing removes pores entirely, creating that waxy look. This prompt tells Nano Banana 2 to reduce blemishes while keeping the skin’s natural surface structure intact.
12. Selfie to Studio Portrait
When to use it: A front-camera selfie that needs to look like it was taken by a professional photographer in a studio.
Transform this selfie into a professional photoshoot-quality image by improving lighting, adding soft background blur, and enhancing facial sharpness. Balance skin tones, refine exposure, and ensure the subject stands out clearly. Create a clean, visually appealing result that resembles studio photography.
Why it works: The “soft background blur” instruction simulates the shallow depth of field you get from expensive portrait lenses. Combined with “studio photography” as the quality target, this bridges the gap between a phone selfie and a $500 headshot.
13. Enhance Facial Details Naturally
When to use it: Portraits that are decent but lack the crisp detail in eyes, eyebrows, and facial features that makes a photo feel alive.
Improve facial details by enhancing natural skin texture, refining highlights and shadows, and increasing clarity. Avoid over-smoothing while ensuring a polished appearance. Adjust lighting to highlight key features and create a balanced, professional portrait that feels realistic and visually appealing.
Why it works: “Avoid over-smoothing” and “natural skin texture” are guardrails that keep the AI honest. This prompt enhances rather than replaces — it’s the difference between touching up a portrait and painting over it.
14. Add DSLR-Style Background Blur
When to use it: Your subject blends into a busy background. You need that creamy bokeh separation without reshooting.
Add depth of field by softly blurring the background while keeping the subject in sharp focus. Simulate a DSLR lens effect with smooth bokeh and natural transitions. Enhance subject clarity and lighting, ensuring the final image looks professional and visually separated from the background.
Why it works: “Smooth bokeh and natural transitions” prevents the hard-edge cutout look that cheap portrait mode creates. Real lens blur has a gradient — sharp at the subject, progressively softer as distance increases. This prompt asks for that physics-accurate falloff.
15. Magazine-Cover Portrait
When to use it: Professional headshots, portfolio images, or any portrait that needs to look publication-ready.
Convert this portrait into a magazine-style image by refining lighting, enhancing contrast, and balancing tones. Sharpen key features, improve composition clarity, and create a premium editorial look. Ensure the result feels polished, high-end, and suitable for professional publication or branding.
Why it works: “Editorial look” and “professional publication” set a specific quality bar. Nano Banana 2 interprets this as the particular combination of precise lighting, controlled contrast, and subtle color grading that distinguishes magazine imagery from everyday photography.
Category 4: Backgrounds and Scene Changes
Nano Banana 2’s biggest leap over previous models is scene manipulation. It can isolate subjects, rebuild environments, and blend everything so naturally that the edit becomes invisible. These prompts exploit that strength.
16. Remove and Replace Background
When to use it: Your subject is great but the background is distracting — a cluttered room, a parking lot, someone photobombing.
Remove the existing background and replace it with a clean, professional studio backdrop using soft neutral gray tones. Maintain the subject's original lighting direction and color temperature. Ensure seamless edge blending with no visible cutout artifacts, and preserve all fine details like hair strands and clothing edges.
Why it works: “Maintain the subject’s original lighting direction” is the detail that separates convincing composites from obvious fakes. If your subject was lit from the left, the new background needs to match. This prompt makes that explicit.
17. Add Landscape Depth
When to use it: Landscape and travel photos that look flat on screen compared to how the scene felt in person.
Enhance this landscape by adding atmospheric depth and dimension. Apply subtle haze to distant elements, increase foreground sharpness, and create natural tonal separation between foreground, midground, and background layers. Boost color vibrancy slightly while maintaining realistic tones and a natural outdoor feel.
Why it works: Real landscapes have atmospheric perspective — distant objects look hazier and bluer. Phone cameras compress this. The prompt reconstructs three distinct depth layers, mimicking how human eyes actually perceive distance.
18. Change Season or Weather
When to use it: You need a winter version of a summer photo, autumn leaves in a spring scene, or rain on a sunny day.
Transform the environment in this photo to reflect autumn by changing foliage to warm red, orange, and gold tones. Adjust the overall lighting to match soft, diffused overcast conditions typical of fall. Maintain the subject and foreground elements exactly as they are while ensuring the seasonal change looks seamless and realistic throughout the background.
Why it works: Swap “autumn” for any season you need. The key is specifying both the visual change (foliage color) AND the corresponding light change (overcast vs. harsh sun). Seasons affect everything — light quality, shadow softness, color temperature — and this prompt accounts for all of it.
19. Create a Studio Product Backdrop
When to use it: Product photos shot on a kitchen counter or office desk that need to look like professional product photography.
Place this product on a clean, minimalist surface with a smooth gradient background that transitions from white to light gray. Add soft, even studio lighting with gentle reflections beneath the product. Remove all surrounding clutter and distractions while keeping the product's colors, textures, and proportions completely accurate.
Why it works: “Gentle reflections beneath the product” creates that polished surface look you see in Apple product shots. The gradient background adds depth without distraction. And “completely accurate” colors prevent the AI from color-correcting your product into something it isn’t.
20. Teleport Your Subject to a New Location
When to use it: Travel content, creative composites, or any time you want your subject in a location they never visited.
Place this person in a bustling Tokyo street scene at night with neon signs and ambient city glow. Match the lighting on the subject to the environment — cool blue and pink neon reflections on skin and clothing. Maintain the subject's exact appearance, facial features, and outfit while ensuring the scene integration looks completely natural with proper shadows and scale.
Why it works: Nano Banana 2 excels at character consistency — keeping your face identical across edits. The prompt asks for environment-matched lighting (neon reflections on skin), which is the detail that sells the illusion. Swap “Tokyo street scene” for any location you need.
Category 5: Restore and Upscale
Old family photos. Damaged prints. Low-resolution images from a decade ago. Nano Banana 2 can reconstruct detail that doesn’t exist in the original file — not by guessing, but by understanding what faces, textures, and objects should look like at higher resolution.
21. Restore Faded Old Photos
When to use it: Scanned family photos that have yellowed, faded, or lost contrast over the decades.
Restore this old photo by removing noise, fixing faded areas, and enhancing details carefully. Improve clarity while preserving the original character of the image. Adjust colors subtly to bring back life, ensuring a balanced and natural restoration result.
Why it works: “Preserving the original character” is the instruction that stops the AI from turning your grandmother’s 1960s portrait into a modern iPhone photo. You want it cleaner, not different. The prompt respects the era while fixing the damage.
22. Colorize Black-and-White Images
When to use it: Vintage black-and-white photos that you want to see in realistic, historically accurate color.
Colorize this black-and-white image using realistic and natural tones. Ensure accurate skin colors, balanced lighting, and smooth transitions. Maintain historical authenticity while enhancing clarity and detail to create a visually compelling and lifelike final image.
Why it works: “Historical authenticity” tells Nano Banana 2 to consider the era of the photo when choosing colors. A 1940s portrait gets different clothing and environmental tones than a 1980s snapshot. The AI draws on its training data to make period-appropriate color choices.
23. Upscale to 4K Resolution
When to use it: Low-resolution images — old digital photos, screenshots, or compressed web images — that need to be printed or displayed at a larger size.
Upscale this image to 4K resolution while enhancing sharpness and refining textures. Avoid artificial artifacts, maintain natural detail, and improve overall clarity. Ensure the final result looks clean, realistic, and suitable for high-quality digital or print use.
Why it works: Standard upscaling just stretches pixels. AI upscaling reconstructs detail. The “avoid artificial artifacts” instruction prevents the hallucinated textures that cheaper upscaling tools produce — those weird patterns that appear on skin and fabric when the AI fills in pixels it isn’t sure about.
24. Repair Scratched or Damaged Photos
When to use it: Physical prints that were scratched, water-damaged, folded, or partially destroyed before scanning.
Repair damaged areas such as scratches, spots, or missing details by reconstructing them naturally. Enhance clarity and improve overall image quality while maintaining authenticity. Ensure seamless blending so the restoration looks clean and realistic.
Why it works: “Reconstructing them naturally” and “seamless blending” tell the AI to inpaint damaged areas using context from the surrounding image. Nano Banana 2 doesn’t just smear pixels — it understands what a face, a wall, or a tree should look like and rebuilds the missing pieces accordingly.
25. Enhance Vintage Photos Without Killing the Vibe
When to use it: Old photos where you want better quality but don’t want to lose the vintage character that makes them special.
Enhance this vintage photo while preserving its original aesthetic and mood. Improve sharpness, refine colors subtly, and maintain texture. Ensure the final image feels authentic yet visually improved, balancing restoration with originality.
Why it works: The tension between “enhance” and “preserve” is the point. You’re giving the AI two competing instructions and forcing it to find the middle ground — sharper, but still vintage. Cleaner, but still warm. That balance is exactly what you want.
Category 6: Social Media Ready
These prompts optimize for the specific contexts where your photos will be viewed — phone screens, feed scrolling, thumbnail grids. Each one targets a different platform behavior.
26. Instagram-Optimize Any Photo
When to use it: Any photo destined for your Instagram feed that needs to look polished without looking edited.
Enhance this image for Instagram by slightly boosting colors, improving sharpness, and refining lighting. Maintain a natural yet aesthetic tone, ensuring the image feels clean, modern, and visually appealing without looking overly edited.
Why it works: “Slightly” and “without looking overly edited” are doing the heavy lifting. Instagram’s algorithm favors high-engagement images, and over-edited photos consistently underperform natural-looking ones. This prompt adds just enough polish to stop the scroll without triggering the “that’s fake” reflex.
27. Perfect Profile Picture
When to use it: LinkedIn, Twitter, WhatsApp — anywhere a small circular crop needs to make a strong impression.
Optimize this photo for a profile picture by sharpening facial features, cleaning the background, and improving lighting balance. Ensure high clarity and focus on the subject, creating a strong and professional visual identity.
Why it works: Profile pictures are viewed at tiny sizes. “Sharpening facial features” and “cleaning the background” maximize the visual signal in a very small space. Every pixel matters when your face is a 100px circle.
28. Stop-the-Scroll Feed Images
When to use it: Any image that needs to compete for attention in a fast-moving social media feed.
Create a scroll-stopping image by increasing contrast, enhancing colors, and sharpening the subject. Improve clarity and visual depth while maintaining balance, ensuring the image stands out immediately in social media feeds.
Why it works: Feed images are surrounded by competing content. Higher contrast and enhanced color saturation create visual weight that pulls the eye. “Stands out immediately” tells the AI to optimize for first-glance impact — the 0.3 seconds you get before someone’s thumb keeps scrolling.
29. YouTube Thumbnail That Gets Clicked
When to use it: Video thumbnails that need to pop in a grid of competing content at small display sizes.
Transform this image into a YouTube thumbnail style by increasing brightness, boosting contrast, and enhancing sharpness. Make the subject highly visible and attention-grabbing, ensuring the image performs well in high-competition environments.
Why it works: YouTube thumbnails live or die at approximately 320 × 180 pixels. At that size, subtlety is invisible. This prompt cranks the key variables — brightness, contrast, sharpness — to maximum effectiveness for small-format viewing.
30. Influencer-Quality Aesthetic
When to use it: Lifestyle, fashion, or brand content that needs that curated, high-production-value look.
Make this image look like influencer-style content by applying soft lighting, premium tones, and a polished aesthetic. Enhance clarity, refine colors, and create a visually appealing composition that feels modern and high-end.
Why it works: “Influencer-style” is a legitimate aesthetic category that Nano Banana 2 recognizes — soft shadows, slightly desaturated highlights, warm mid-tones, and an overall sense of effortless polish. Naming the style directly is more effective than describing each individual adjustment.
What’s Actually Happening Under the Hood
Nano Banana 2 launched on March 19, 2026 — just days ago. It combines the studio-grade intelligence of Nano Banana Pro with the speed of Gemini Flash, making every prompt in this list execute faster than the previous generation could manage.
The model understands real-world physics. It knows how light wraps around a face, how shadows fall on fabric, how atmospheric haze works at distance. When you say “golden hour lighting,” it doesn’t apply an orange overlay — it simulates the actual behavior of low-angle sunlight.
Google also added new markup tools in beta this month. You can now circle a specific area of your image and type an edit instruction directly — no more switching between the editor and the chat. For these prompts, that means you could circle just the sky and run a golden hour edit on it while leaving the foreground untouched.
The AI photo editing market is projected to more than double by 2033, reaching $2.2 billion — and over 58% of independent creators already use AI-assisted editing as their primary production method. This isn’t early-adopter territory anymore. This is the new default.
How to Build Your Own Style
These 30 prompts are starting points, not endpoints. The real skill is combining them.
Run Prompt 2 (low-light fix) first, then Prompt 8 (Hollywood grade) second. Fix the technical problems, then apply the creative vision. That two-pass approach consistently produces better results than trying to do everything in a single prompt.
Once you find combinations you like, save them. Build a personal library of three to five stacked prompt sequences that define your visual brand. Consistency beats perfection — an Instagram feed with a recognizable look outperforms one where every photo was edited differently, even if some individual edits are stronger.
One more thing: Nano Banana 2 adds SynthID watermarks to edited images — both visible and invisible. This is Google’s approach to AI content identification, and it applies to every edit you make through the Gemini app. Keep this in mind for commercial use and always disclose AI editing where platform policies or clients require it.