Free AI Tool • 1 free generation • Sign-up optional

Nano Banana Image-to-Image AI Generator

Edit any photo with plain English: style transfer, background change, object tweaks — in seconds.

~10–15s per generation
1 free generation (no sign-up)

Try 1 free generation — sign up optional. Sign up to get 3 more free credits.

Try Image-to-Image Now

Upload your image and describe the transformation you want. Try 1 free generation — sign up optional.

0/1000

Popular styles

No sign up required for your first try • Sign up for 3 more credits

Your image will appear here

Upload an image and describe your changes to get started

Want more features? Try the full editor

    See Nano Banana Img2Img in Action

    Style

    Cinematic Variation

    Before
    Original street fashion photo before AI transformation
    After
    Cyberpunk cinematic variation with neon lighting
    Prompt:

    Cinematic night scene with neon city reflections. Keep the same person, pose, and composition. Add dramatic lighting with cyan and magenta tones. Wet street reflections, bokeh lights in background. No text.

    Background

    Background Swap

    Before
    Original photo with urban background
    After
    Same subject with sunny beach boardwalk background
    Prompt:

    CRITICAL: Keep the subject EXACTLY the same (face identity, hair, skin tone, expression, body proportions, pose/walking posture, clothing, bag, and camera framing). Do not change the subject at all. ONLY CHANGE: Replace the background with a sunny beach boardwalk scene: wooden boardwalk leading into the distance, palm trees, bright blue sky, ocean visible, warm natural sunlight. LIGHTING MATCH: Ensure the subject lighting matches the beach scene (same sun direction, color temperature, and shadow strength). Add a subtle natural ground shadow consistent with the boardwalk. EDGE BLENDING: Clean, realistic edges around hair and clothing (no cutout look, no halos). Keep fine hair strands. DEPTH: Match depth-of-field so subject and background feel like the same photo (consistent blur and perspective). No text, no watermark, no extra people added.

    Character

    Outfit Change

    Before
    Original photo with casual outfit
    After
    Same person with quiet luxury old money outfit
    Prompt:

    CRITICAL: Keep the subject EXACTLY the same (face identity, hair, skin tone, expression, body proportions, pose, and camera framing). Keep the background EXACTLY the same (same beach boardwalk, palm trees, sky, ocean). Keep the bag EXACTLY the same. ONLY CHANGE: Replace the clothing with a quiet luxury "old money" outfit: camel structured coat, cream premium knit sweater, navy tailored trousers. Elegant and minimal. No new accessories added. Photorealistic fabric texture with natural folds. Seamless transitions at neckline, sleeves, and waist. Match the original daylight direction and shadow softness. No text, no watermark.

    Product

    Product White BG

    Before
    Product photo with original background
    After
    Product on pure white studio background
    Prompt:

    CRITICAL: Keep the product EXACTLY the same (shape, design, color, materials, laces, stitching). Do NOT change the sneaker design in any way. ONLY CHANGE: Remove the background completely and place the product on a pure white seamless studio background (#FFFFFF). No text, no watermark.

    Quick Answer

    Nano Banana Img2Img (image-to-image) lets you edit a photo using a text prompt while preserving key details like identity, pose, and composition. Try 1 free generation — sign up optional. Sign up to get 3 more free credits.

    What Nano Banana Img2Img excels at

    Powered by Gemini 2.5 Flash Image, Nano Banana handles these img2img tasks exceptionally well:

    Character Consistency

    Keep the same face, hair, and identity while changing style, outfit, or background.

    Learn more

    Combine Photos

    Merge elements from multiple images into a single coherent scene with matched lighting.

    Learn more

    Local Edits

    Change specific parts (outfit, background, objects) without affecting the rest of the image.

    Learn more

    How to use (3 steps)

    1

    Upload a photo

    Clear subject, good lighting, 1024x1024+ recommended for best results.

    2

    Paste a recipe prompt

    Copy one of the recipes below and adjust the variables (style, background, outfit).

    3

    Generate

    Click Generate. If the result is off, use the troubleshooting tips below to refine.

    Prompt Recipes

    Copy a recipe, replace the [brackets] with your specifics, and generate.

    Character

    Keep identity, change outfit

    CRITICAL: preserve the same face, hair, skin tone, pose, background. ONLY change clothing to: [new outfit]. Photorealistic fabric texture, consistent lighting, seamless transitions.
    Background

    Background swap (keep subject)

    Preserve the subject exactly. Replace only the background with: [new background]. Match lighting direction and color temperature. No artifacts around edges.
    Style

    Style transfer (subtle)

    Keep identity and composition. Apply a subtle [style] look: [style notes]. Maintain realistic skin texture and natural shadows.
    Style

    Cinematic variation

    Cinematic lighting, shallow depth of field, realistic textures. Keep subject unchanged and preserve composition. No text.
    Product

    Product photo (white background)

    Keep the product exactly the same. Remove the background and place product on a pure white background. Preserve edges and true colors. No text, no watermark.
    Composite

    Combine two photos

    Use Image 1 as the base. Add [object/person] from Image 2 naturally. Match lighting, scale, and perspective. Seamless integration.
    Environment

    Seasonal transformation

    Transform to [season] scene with appropriate colors and lighting. Keep subject and composition unchanged. Natural atmosphere.
    Interior

    Interior/Room redesign

    Keep the room layout and perspective. Change furniture style to [style]. Match lighting and maintain realistic proportions.

    Common issues and fixes

    1

    Result doesn't look like the original person

    Fix: Add stronger constraints: "EXACT same face, hair, skin tone, pose"

    2

    Background changed when you only wanted outfit change

    Fix: Add "DO NOT modify background" and specify "ONLY change [specific thing]"

    3

    Edges look messy or have halos

    Fix: Add "clean edges, no halos, no cutout artifacts, seamless transitions"

    4

    Lighting doesn't match

    Fix: Add "match original light direction and color temperature"

    Image-to-image vs image editing

    Both features are available in Nano Banana, but they serve different purposes.

    Use Editing For:

    • Precise changes: remove a specific object
    • E-commerce: pure white backgrounds
    • Corrections: fix lighting, remove blemishes
    • When subject must stay exactly identical

    Use Image-to-Image For:

    • Creative variations: different styles, moods
    • Dramatic transformations: cinematic looks
    • Exploring options: multiple variations
    • When you want AI to be more creative

    Need precise edits like removing backgrounds or objects? See AI Image Editing prompts (remove background & objects) →

    Frequently Asked Questions

    Quick answers to common questions about Nano Banana image-to-image

    Image-to-image (img2img) uses your uploaded photo as a starting point to generate new variations. Unlike text-to-image which creates from nothing, img2img preserves elements you specify while transforming others — style changes, mood shifts, background replacement, and more.

    Image editing makes precise, controlled changes (remove object, exact background color). Image-to-image creates artistic variations and transformations. Use editing when you need exact control; use img2img when you want creative exploration. Both are available in Nano Banana.

    Include explicit constraints in your prompt: 'keep the same person exactly', 'preserve identity and pose', 'maintain original composition'. The more specific you are about what should stay the same, the better. For portraits, also add 'no face changes' and 'natural skin texture'. Use fewer constraints for more creative freedom.

    Use clear, well-lit images with the subject prominently visible. Resolution of 1024x1024 or higher works best. Avoid heavily compressed images, blurry photos, or obscured details. Quality in means quality out.

    Most transformations complete in 10-15 seconds using Gemini 2.5 Flash Image (Nano Banana). Complex variations may take slightly longer.

    Yes! Try 1 free generation — sign up optional. Sign up (email only, no credit card) to get 3 more free credits. That's 4 total free transformations to test the feature.

    Yes, all images you generate can be used for commercial purposes. This includes variations created from your own photos, marketing materials, social media content, and client work. You retain full usage rights.

    API access is not offered at this time. Nano Banana is available through the web interface only.

      Free Nano Banana Img2Img | Image-to-Image AI Tool