Photo for Author | Gemini (Nano-Banana Portrait)
Obvious Introduction
Angi-Geni Venerative AI has been widely used for both businesses and businesses, allowing them to cause immediate viewing without requiring any design technology. In fact, these tools can speed up jobs that would take some important time, to complete it in seconds.
With technological and competitive development, many modern, improved, productive products are issued, similar to Stable importation, Elder, Dall-e, Promiseand many others. Each offers different benefits from its users. However, Google has just made a major impact on the example of the image production in the release of Gemini 2.5 Flash image (or nano-banana).
Nano-banana model for the Google and editorial image, with skills such as creating a realistic image, a mix of photos, letters, targeted public transformation, and public achievement. The model offers more control over past models from Google or its competitors.
This document will evaluate the power of developing and banana production. We will show these features using Google Ai Studio platform with Gemini API within the nature of Python.
Let's get into it.
Obvious To check the Nano-banana model
Following this lesson, you will need to subscribe to the Google Account and log in to Google Ai Studio. You will also need to find an API key To use Gemini API, you need a paid program as there is no free conference available.
If you choose to use API via Python, be sure to install Google Orandatory AI Library for the following command:
When your account is set, let's check how to use the Nano-Banana model.
First, show it to Google Ai Studio and select Gemini-2.5-flash-image-preview Model, the Nano-Banana model will be using it.
With the selected model, you can start a new conversation to produce a picture from immediately. As Google suggests, the basic goal of getting the best results Describe the incident, not just a list of keywords. This account, describing the picture you see, usually produces higher results.
In the AI Studio chat conversation, you will see a platform similar to the below where you can apply immediately.
We will use the following to produce photo photoelistic image by our example.
Fixed photo of Indonesia Batik Artisan, colored hands with wax, tracking flowing motif with indigo cloth with cleaning pen. She works at a wooden table in the windy veranda; Folded fabrics and dye of vies blossom behind him. The light of the window is late in all the fabric, producing good wax lines and teak breasts. It is included in 85 mm in F / 2 with gentle separation and creamy bocheh. The perfect weather focuses on, unclean, and proud.
The product produced is shown below:
As you can see, the image made of facts and adheres to the achievement given. If you select Psytho Use, You can use the following code to create a picture:
from google import genai
from google.genai import types
from PIL import Image
from io import BytesIO
from IPython.display import display
# Replace 'YOUR-API-KEY' with your actual API key
api_key = 'YOUR-API-KEY'
client = genai.Client(api_key=api_key)
prompt = "A photorealistic close-up portrait of an Indonesian batik artisan, hands stained with wax, tracing a flowing motif on indigo cloth with a canting pen. She works at a wooden table in a breezy veranda; folded textiles and dye vats blur behind her. Late-morning window light rakes across the fabric, revealing fine wax lines and the grain of the teak. Captured on an 85 mm at f/2 for gentle separation and creamy bokeh. The overall mood is focused, tactile, and proud."
response = client.models.generate_content(
model="gemini-2.5-flash-image-preview",
contents=prompt,
)
image_parts = [
part.inline_data.data
for part in response.candidates[0].content.parts
if part.inline_data
]
if image_parts:
image = Image.open(BytesIO(image_parts[0]))
# image.save('your_image.png')
display(image)If you give your API key and Prepred to want, the Python code above will generate a picture.
We have seen that model Nino-banana can produce photo picture. As mentioned earlier, Nano-banana has great power to edit the image, which will be evaluated next.
Let's try to organize a scroll based on a newly produced image. We will use the next time to change the appearance of Artisan's appearance:
Using a given photo, place pair of tiny glasses for gently learning at Artisan nose while dragging wax lines. Make sure the reminders look realistic and mirrors live naturally in his face without hiding his eyes.
The image that appears is shown below:
The image above is the same as the first, but by mirrors are added to the surface of Artisan. This shows how Nano-banana can plan to organize the picture according to speeding while storing complete harmony.
To do this with Python, you can give your basic picture and the new TEMPT using the following code:
from PIL import Image
# This code assumes 'client' has been configured from the previous step
base_image = Image.open('/path/to/your/photo.png')
edit_prompt = "Using the provided image, place a pair of thin reading glasses gently on the artisan's nose..."
response = client.models.generate_content(
model="gemini-2.5-flash-image-preview",
contents=[edit_prompt, base_image])Next, let's examine the alteration of characters by generating a new situation when the art is looking directly to the camera and smile:
Generate new photo and image using a given image as a reference to ID: The same batik artisan is now looking at the camera with renewable smile. Medium, 85 mm Appearance and soft Veranda light, direct pots dehydrated.
The image effect is displayed below.
We have changed this scene while keeping the alteration of characters. Exploring the biggest change, let us use the following to see how Nano-Banana made it.
Create a picture in the product style using a given image as an ID reference: a similar scholar introduces the completed Batik Fabric, arms expanded to the camera. It is soft, even light in the window, a look at 50 mm, neutral clutter.
The result is displayed below.
The image that appears shows a completely different location but keeps the same character. This highlights the ability of the model to produce facts that are different content from one reference.
Next, let's try the photo style transmission. We will use the next time to change the photorealistic image into a watercolor water drawing.
Using a given photo as ID reference, multiply a place as a cold watercolor on the cold paper keep your poster with a fabric, gentle smile and circular glasses; Allow the Veranda down to a bright ninth and a decrease in visual paper.
The result is displayed below.
The picture shows that the style has been converted into the watercolor while storing the title and original formation.
Finally, we will try the photos, where we put the item from one picture to another. For example, I have produced a picture of a woman's hat using Nano-banana:
Using a picture of the hat, we will now put the head of Artisan at the next time:
Submit the same woman and put outside the open shade and place the straw hat on the product picture. Mix the crown and the head of a real head; Brief on her right ear (left camera), the ribbon tails drove slowly with gravity. Use soft sky light as a keys to a gentle line from the bright behind. Stay the true grass and grass grass and natural leather ton, and a convincing shadow from the forehead and over glasses. Keep batik cloth and his hands unchanged. Keep the watercolor style of watercolor unchanged.
This process includes an image of the hat with the basis of the foundation to produce a new image, with small changes in the pose and the entire style. In Python, use the following code:
from PIL import Image
# This code assumes 'client' has been configured from the first step
base_image = Image.open('/path/to/your/photo.png')
hat_image = Image.open('/path/to/your/hat.png')
fusion_prompt = "Move the same woman and pose outdoors in open shade and place the straw hat..."
response = client.models.generate_content(
model="gemini-2.5-flash-image-preview",
contents=[fusion_prompt, base_image, hat_image])For the best results, use a higher number of three installations. Much use can reduce the quality of the output.
That includes the basics of using the Nano-banana model. In my opinion, this model passes when you have the pictures there you want to change or edit. It is especially useful at the end of the setting in a series of pictures produced.
Try yourself and don't be afraid to deteriorate, as you often have nothing to do in the first attempt.
Obvious Rolling up
Gemini 2.5 Flash image, or nano-banana, is the latest generation generation from Google. Has strong skills in comparison with modern generation status models. In this article, we checked how we can use Nano-banana production and edit photos, highlighting its features to maintain agriculture and use the stylistic.
I hope this is helpful!
Cornellius Yudha Wijaya It is a scientific science manager and the database author. While working full-time in Allianz Indonesia, she likes to share the python and data advice with social media and media writing. Cornellius writes to a variety of AI and a study machine.
