Looking for the best Google Gemini AI photo editing prompts? This guide includes 15 copy-and-paste Nano Banana prompts for portraits, professional headshots, old photo restoration, background replacement, outfit changes, cinematic edits, product photography, and more.

Whether you’re using the free version of Gemini or Nano Banana Pro, these prompts are designed to preserve facial identity, improve realism, and produce consistent, high-quality edits. Instead of vague one-line prompts, you’ll find detailed prompts that tell Gemini exactly what to change – and what to leave untouched.

Every prompt has been organized by use case, so you can quickly findthe right one, copy it, customize it, and start editing withinminutes.

How to Use Google Gemini for AI Photo Editing

As of 2026, Gemini’s default image model is Nano Banana 2 (Gemini 3.1 Flash Image), with Nano Banana Pro (Gemini 3 Pro Image) available as a higher-fidelity option for complex compositions, accurate multi-language text, and up to 4K output. You access it from the “Create images” tool inside the Gemini app, choosing Fast, Thinking, or Pro mode, then upload a photo and describe your edit. Every image Gemini creates or edits carries a visible watermark plus Google’s invisible SynthID marker.

One detail worth knowing upfront: image editing in the Gemini app is restricted to users 18 and older, even though basic image generation is available more broadly, so these prompts assume an adult uploading and editing their own photos.

How to Use Gemini Photo Editing Prompts Effectively

  • Open the Gemini app, tap the tools menu, and select “Create images.”
  • Pick Fast for quick everyday edits, Thinking for more complex multi-step requests, or Pro for the highest fidelity and best text rendering.
  • Upload your photo (or two photos, if you’re blending), then paste your prompt.
  • If you want to swap models on a result you already generated, use the three-dot menu and “Redo with Pro” for a higher-quality re-run.
  • Refine in follow-up messages rather than rewriting the whole prompt – Nano Banana is built to hold onto a subject’s likeness across an editing conversation.

The Ultimate AI Photo Prompt Formula for Google’s Gemini/Nano Banana

The most consistent results come from telling Gemini the subject, the specific change, what has to stay identical, and the mood or style you’re after – in that order.

Weak prompt:

“Make this photo cooler.”

Better prompt:

“Restyle this photo with a 90s film grain aesthetic – warm faded tones, soft grain, slightly lifted shadows. Keep the subject’s facial identity, expression, hairstyle, and clothing exactly as they appear in the original. Photorealistic result, not illustrated.”

The difference isn’t length for its own sake – it’s that the second version gives Gemini a subject, a clear stylistic target, and an explicit boundary on what not to touch.

Best AI Portrait and Selfie Editing Prompts

Natural Portrait Touch-Up

Best for: a cleaner version of a selfie or portrait without an obvious “AI” look

Prompt:

Improve the lighting and clarity of this portrait – even out exposure, correct white balance, reduce harsh shadows. Keep facial identity, skin texture, expression, and proportions exactly as they are in the original. Do not smooth skin or reshape any features. Photorealistic output only.

Portrait Prompt for Him

Best for: a defined, editorial-style portrait for a masculine-presenting subject

Prompt:

Adjust this portrait with cooler-toned, higher-contrast lighting for a sharper, editorial look. Keep facial structure, skin texture, expression, and hairstyle unchanged, sharpen detail naturally without smoothing. Preserve the original clothing and pose. Photorealistic result.

What it does: leans into contrast rather than softness, which reads as a more defined, magazine-style edit.

Advanced AI Background Swap and Photo Blending Prompts

This is where Nano Banana genuinely stands apart, it can combine two separate photos into one coherent scene while keeping each subject’s likeness intact.

Combine Two Photos

Best for: putting two people (or a person and a pet) into the same scene from separate photos

Prompt:

Combine these two photos into a single natural scene: [describe the setting, e.g. “a sunlit park bench”]. Keep both subjects’ facial identity, proportions, and clothing exactly as shown in their original photos. Match lighting direction and color temperature across both subjects so the scene looks like it was shot in one take.

Background Replacement

Best for: swapping out the setting behind a subject

Prompt:

Replace the background of this photo with [describe new background]. Match lighting, shadow direction, and color temperature on the subject to the new setting. Keep the subject- face, hair, clothing, pose – completely unchanged, with clean, natural edges.

Professional AI Headshot and LinkedIn Photo Prompts

Corporate Headshot

Best for: converting a casual photo into a usable professional headshot

Prompt:

Turn this into a professional headshot: plain neutral studio background, soft even lighting, subtle natural smile if the original allows it. Keep facial identity, skin texture, and proportions exactly as in the original. Photorealistic, not stylized.

Mini Figurine Style

Best for: the collectible-figurine trend that made Nano Banana popular in the first place

Prompt:

Turn this photo into a realistic collectible figurine on a display stand, with the subject’s likeness, outfit, and pose preserved as closely as possible in miniature form. Add a simple display base and soft studio lighting behind it.

Retro Studio Portrait

Best for: 90s or 80s-style throwback portraits for social media

Prompt:

Turn this into a retro-style studio portrait [choose an era, e.g. “90s mall photo studio,” “80s preppy”] backdrop, period-appropriate color grading and lighting. Keep facial identity, expression, and proportions unchanged underneath the style treatment.

Virtual Outfit and Hairstyle Change Prompts

Outfit Change

Best for: previewing a different outfit on an existing photo

Prompt:

Change the outfit to [describe outfit]. Match lighting and shadow on the new clothing to the rest of the photo. Keep face, hair, pose, and background exactly as they are.

Hairstyle Preview

Best for: trying a new haircut or color before committing

Prompt:

Change the hairstyle to [describe style, e.g. “shoulder-length waves” or “short fade”]. Keep facial features, skin tone, expression, and the rest of the photo completely unchanged. Make the hair color and texture look natural, not painted on.

Couple Photo Editing and Romantic AI Restoration Prompts

Romantic Golden-Hour Edit

Best for: giving an existing couple photo a warmer, more intentional feel

Prompt:

Enhance this couple photo with warm golden-hour lighting, soft contrast, and a gentle cinematic color grade. Keep both people’s facial identity, expressions, and poses exactly as they are in the original. Do not alter body positioning or add elements that weren’t in the photo.

Anniversary / Wedding Photo Restoration

Best for: cleaning up and enhancing an older couple or wedding photo

Prompt:

Restore and enhance this photo – correct fading and color shift, improve sharpness and lighting balance. Keep both subjects’ facial identity, expressions, and clothing details historically accurate to the original. Do not invent details that aren’t visible in the source image.

Tip: for combining separate individual photos into one romantic scene, use the two-photo blending prompt above and describe a shared setting – it tends to produce a more natural result than asking Gemini to “add a partner” to a single photo from imagination.

Cinematic Lighting and Color Grading AI Prompts

Cinematic Color Grade

Best for: adding mood and depth to a flat photo

Prompt:

Apply a cinematic color grade to this photo – richer shadows, warm highlights, subtle contrast boost, shallow depth of field on the background. Keep the subject’s facial features, expression, and clothing exactly as they are.

How to Restore Old Photos with AI: Best Prompts

Damage Repair & Colorization

Best for: scanned prints with scratches, fading, or no color

Prompt:

Restore this old photo: repair scratches, tears, and faded areas, correct discoloration, and improve sharpness. If the photo is black and white, add natural, era-appropriate color. Keep facial features, expressions, and clothing exactly as shown, do not invent details not visible in the original.

Product Photography and Infographic AI Prompts for Nano Banana Pro

Nano Banana Pro’s real-world grounding and improved text rendering make it noticeably better than the base model for anything involving legible text, diagrams, or factual accuracy.

Product Shot on Clean Background

Best for: turning a casual product photo into something closer to catalog-ready

Prompt:

Place this product on a clean white background with soft, even studio lighting and a subtle natural shadow. Keep the product’s shape, color, texture, and branding exactly as in the original, do not alter labels or text.

Infographic From Notes

Best for: turning handwritten notes or a rough sketch into a clean visual

Prompt:

Convert this into a clean infographic with clear, accurate text matching what’s written in the original notes, organized headings, and a simple, readable layout. Use [describe color scheme]. Keep all numbers and labels factually identical to the source.

How to Get Better Results with Gemini

  • Choose Pro mode when text accuracy or complex composition matters more than speed.
  • When blending two photos, describe the shared setting explicitly rather than leaving it to Gemini to guess where the scene takes place.
  • Name lighting direction and color temperature instead of just saying “better lighting.”
  • Keep the preservation clause in every prompt involving a real person’s face – it’s the difference between an edit and a different-looking person.
  • Refine within the same conversation instead of starting a new chat each time; Nano Banana is designed to hold context and likeness across follow-ups.
  • Expect the visible and SynthID watermarks on every output – they’re not removable and that’s intentional.

Common AI Photo Editing Mistakes to Avoid

The most common one is treating Gemini like a search engine and typing a single vague noun phrase “cinematic photo” instead of a full instruction. Skipping the preservation clause is a close second; without it, even a simple color grade can drift into a subtly different face. People also frequently forget that editing is restricted to adult accounts, then get confused when a request is declined for reasons that have nothing to do with the prompt itself. And when blending two photos, not specifying the setting or lighting match tends to produce a result where the two subjects look like they were pasted next to each other rather than photographed together.

Getting Started with Gemini AI Image Editing

Pick the category that matches the photo actually sitting on your phone right now, not the one that sounds most impressive. A restoration prompt on a scratched-up family photo or a clean background swap on a headshot will teach you more about how Nano Banana responds to your wording than any amount of reading about it. Once you’ve got a result you like, the fastest way to improve it further is a short follow-up prompt naming the one thing that’s still off – not a full rewrite from scratch.

Frequently Asked Questions
Is Nano Banana free to use?

Basic access is available in the free Gemini tier with usage limits. Google AI Pro, Plus, and Ultra subscribers get higher limits and access to Nano Banana Pro’s advanced features.

What's the actual difference between Nano Banana and Nano Banana Pro?

Nano Banana 2 is optimized for speed and everyday edits. Nano Banana Pro, built on Gemini 3 Pro, is slower but produces higher-fidelity results, more accurate text in multiple languages, and supports up to 4K output — better suited to infographics, detailed compositions, and anything with legible text in it.

Can minors use Gemini's photo editing tools?

No — image editing in the Gemini app is limited to users 18 and older, even though basic image generation has broader availability through supervised, parent-managed accounts.

Are Gemini-edited photos watermarked?

Yes. Every image carries a visible watermark and Google’s invisible SynthID marker, which identifies it as AI-generated or AI-edited.

Can I combine more than two photos at once?

The core blending feature is built around two source images. For more subjects, it’s more reliable to blend in stages — combine two, then bring in the result as one of two inputs for the next blend — than to upload several photos at once and hope Gemini sorts out the composition.