Here's a summary of this tutorial content:
- Use Nano Banana to help you retouch photos: acne removal, body slimming, and face slimming made simple
- Photo rescue tool - use Nano Banana to transform ordinary photos into masterpieces instantly
- Display your daily outfit in a clean and efficient way
- You can also quickly dress your photos in outfits curated by others
- Use marking, boxing, and doodling to help prompts generate images more precisely
- Make full use of Nano Banana's world knowledge to create video effects
- Generate continuous movie storyboards based on character photos with one click
- Turn your favorite photos or items into physical stickers to give to friends
How to Use Nano Banana
First, let's see how to use Nano Banana. I recommend starting with Google's AI Studio (https://aistudio.google.com/) since it's free and gives you the full experience. You can also use it directly in Gemini.
After entering the page, switch the model to Gemini 2.5 Flash Image Preview on the far right side of the page. Yes, this is the official name of Nano Banana.
After switching models, you can input your image modification requirements in the input box while uploading the images you need to modify. Multiple image uploads are supported here. Then click the Run button and wait for the results.
Nano Banana also supports continuous image editing. Note that after four or five consecutive edits, the effect may deteriorate due to context overflow. In this case, I suggest opening a new window and starting fresh.
Besides AI Studio, some Gemini APP users can also use Nano Banana now. If you can use it, you'll receive a similar popup. Just switch to the Gemini 2.5 Flash model and select your image.
Some API service providers and aggregated image/video products we know, like FAL and Krea, have also launched Nano Banana. You can use it on their platforms too. Just don't search for "Nano Banana official website" directly - no such thing exists, and all you'll see are scam sites.
Using Nano Banana for Photo Retouching
Everyone loves taking photos nowadays. Think your photos don't look good? Don't know how to color grade? Got unwanted objects? Leave it to Nano Banana with a simple prompt.
Send it your photos that need work, then use my prompt to transform them into something impressive.
This photo is very boring and plain. Enhance it! Increase the contrast, boost the colors, and improve the lighting to make it richer,You can crop and delete details that affect the composition.
After this upgrade, Nano Banana's facial ID consistency has been greatly enhanced. If you're just making minor modifications to photos, the similarity is very high.
This means we can directly use natural language for photo retouching. For example, making my face slimmer - you can see some facial blemishes were also fixed, so you can directly ask Nano Banana to help fix skin issues.
Make the character's face in the image slimmer, while increasing the muscle mass of the arms.
Nano Banana Helps You Show Off Outfits and Try New Styles
Today's OOTD? Just ask it to display all outfit items in a flat lay.
A flat lay photograph showing all the clothing items involved in the photo.
You can also do the reverse - use your photo to try outfits shared by other bloggers. The reproduction quality is excellent.
When using prompts, pay attention to image order. Here I'm using "first image" and "second image" to indicate.
The character in Figure 2 is wearing the clothing and accessories from Figure 1.
Control Image Modifications Through Doodling and Marking
Since Nano Banana itself has strong multimodal understanding and world knowledge, you can not only modify through text prompts but also directly mark the images that need modification.
Let's start with basic usage. If you have multiple images, you can mark them to help the model better understand your text prompts.
For example, here my prompt can tell it to use the background from image A, the sofa camera angle from image B, and the character from image C to complete a three-image composition.
Using the environment from A, the sofa camera angle from B, and the subject from C, create a new image that combines these three elements.
Taking it further, you can use boxes or circles to mark positions in images that need replacement. Here I had it place the table and chairs from the second image in the red box position.
You can see this is really an e-commerce power tool. The chair details and materials from image 2 are perfectly reproduced, but the lighting and color style perfectly blend into image 1.
Place the chair and table from the first image at the red box location in the second image, and generate the image without the red box markings.
The ultimate doodle control is impressive. You can give it two characters in normal poses plus a rough sketch of their interaction.
It will help you draw interactive images of the two characters in corresponding poses. This is extremely helpful for animation creation and precise action control.
Have these two characters fight using the pose from Figure 3.Add appropriate visual backgrounds and scene interactions,Generated image ratio is 16:9
The above is what I generated with a casually drawn pose. If you can draw your sketches more detailed, Nano Banana's modifications will be more precise.
Like this case from X user minux302, even character expressions were reproduced.
We can expand this further: same character + different action references + continuous editing + Keling 2.1 first and last frames = high-quality fight animation.
The potential here is huge - I'll let you imagine the possibilities.
Turn Your Photos into Cute Stickers
Yesterday I saw Fenx make a bunch of stickers using popular memes from the AI circle and thought it was fun. Since it's open source, I even went to e-commerce platforms to find shops to print a few sets.
Last night after Nano Banana was released, I tried to see if I could generate based on his design reference, and it actually worked.
I also asked about sticker printing - PNG images alone work fine. You can create your own set and find a print shop to make them as gifts for friends. Makes for a thoughtful present.
Help me turn the character into a white outline sticker similar to Figure 2. The character needs to be transformed into a web illustration style, and add a playful white outline short phrase describing Figure 1.
Generate Building or Object Explanation Effects
Google says Nano Banana contains Gemini's world knowledge, so you can add introductions or descriptions of objects in images.
For example, here we can add corresponding AR info cards to images of world-famous buildings. The information about the Taj Mahal and Sydney Opera House here is all correct.
you are a location-based AR experience generator. highlight [point of interest] in this image and annotate relevant information about it.
After getting the images, we process them with first-and-last-frame video models and they directly become explanation animation effects. Here I used Keling 2.1. You can even add corresponding narration audio, and a complete scenic spot introduction storyboard is ready.
E-commerce Image Modification
Previously, FLUX Kontext had an issue with product proportions when modifying e-commerce images, especially with jewelry. Testing Nano Banana here, it performs much better.
Sometimes it changes details in the second image like hairstyle, but you can easily correct this by specifying what to preserve.
The woman in Figure 2 is wearing the necklace from Figure 1,Do not change the details of other Figure 2.
Old Anime Images and Photo Restoration/Upscaling
For the classic challenge of image restoration and upscaling, Nano Banana delivers solid results. Here's an example where I restored an old Ghost in the Shell anime screenshot.
Enhance the resolution of this old anime image and add the appropriate texture details, reinterpreting it with modern anime techniques.
