Exploring Styles and Quality Modifiers in Image Generation

Introduction to Styles and Quality Modifiers

Welcome back! In the previous lesson, you learned how to craft effective prompts by focusing on the key components: subject, context, and style. This lesson builds on that foundation by introducing you to styles and quality modifiers, which are essential for enhancing the quality and artistic expression of your generated images. By the end of this lesson, you will understand how to combine these elements to create visually stunning and high-quality images using Google's Gemini API and its Nano Banana image model.

Understanding Quality Modifiers

Quality modifiers are keywords that enhance the detail and clarity of the generated images. They help the model understand that you are aiming for a high-quality asset. For instance, general modifiers like "high-quality," "beautiful," and "stylized" can be used to improve the overall aesthetic of an image. When generating photos, you might use specific modifiers such as "4K", "HDR" or "Studio Photo" to achieve a more realistic and detailed output. Similarly, for art and illustration, modifiers like "by a professional" or "detailed" can add depth and intricacy to the artwork. To illustrate the impact of quality modifiers, consider the following example. A prompt without quality modifiers, such as "a photo of a corn stalk", might produce a basic image. However, by adding quality modifiers like "4K HDR beautiful photo of a corn stalk taken by a professional photographer", the resulting image will be more vibrant and detailed. It's worth noting that Gemini's image models also let you request specific output resolutions directly through ImageConfig (for example image_size="2K" or "4K"). Resolution keywords in the prompt influence style and detail, while image_size controls the actual pixel dimensions — the two work nicely together.

Exploring Artistic Styles

Artistic styles define the visual approach of the generated image. In this lesson, we will explore two main styles: photography and art/illustration. For photography, you can use prompts that start with "A photo of..." to generate photorealistic images. For example, "A photo of coffee beans in a kitchen on a wooden surface" will produce an image that resembles a real photograph. In contrast, art and illustration styles allow for more creative expression. You can use prompts like "A painting of..." or "A sketch of..." to create images in various artistic styles. For instance, "A technical pencil drawing of an angular sporty electric sedan with skyscrapers in the background" will generate a detailed pencil sketch, while "A digital art of an angular sporty electric sedan with skyscrapers in the background" will produce a vibrant digital artwork.

Join the 1M+ learners on CodeSignal

Be a part of our community of 1M+ users who develop and demonstrate their skills on CodeSignal

Combining Styles with Quality Modifiers: Code Example

Let's dive into a code example to see how you can combine artistic styles with quality modifiers to generate high-quality images. The following code snippet is from app/controllers/image_generator.py.

Python

from google import genai
from google.genai import types
from PIL import Image
from io import BytesIO
import os
import time

# Retrieve API key from system environment variable
api_key = os.getenv("GOOGLE_API_KEY")

if not api_key:
    raise ValueError("GOOGLE_API_KEY not found in environment variables. Set it before running the script.")

base_url = os.getenv("GOOGLE_BASE_URL")

if not base_url:
    raise ValueError("GOOGLE_BASE_URL not found in environment variables. Set it before running the script.")

# Initialize the Gemini client
client = genai.Client(
    api_key=api_key,
    http_options=types.HttpOptions(
        base_url=base_url,
    ),
)

# Define prompts combining artistic styles with quality modifiers
style_quality_prompts = [
    "A beautiful landscape, photorealistic style, 4K resolution",
    "A portrait of a woman, oil painting style, high definition",
    "A futuristic cityscape, cyberpunk style, ultra-realistic, 8K resolution",
    "A serene beach scene, watercolor painting, high-quality details",
]

# Generate images for each combined prompt
for prompt in style_quality_prompts:
    response = client.models.generate_content(
        model="gemini-3.1-flash-image",
        contents=[prompt],
        config=types.GenerateContentConfig(
            response_modalities=["TEXT", "IMAGE"],
        ),
    )
    # Process and display the generated image
    for part in response.parts:
        if part.inline_data is not None:
            image = Image.open(BytesIO(part.inline_data.data))
            timestamp = int(time.time() * 1000)  # Milliseconds precision
            image_filename = f"image_{timestamp}.png"
            output_path = os.path.join("static", "images", image_filename)
            os.makedirs(os.path.dirname(output_path), exist_ok=True)
            image.save(output_path)

from google import genai
from google.genai import types
from PIL import Image
from io import BytesIO
import os
import time

# Retrieve API key from system environment variable
api_key = os.getenv("GOOGLE_API_KEY")

if not api_key:
    raise ValueError("GOOGLE_API_KEY not found in environment variables. Set it before running the script.")

base_url = os.getenv("GOOGLE_BASE_URL")

if not base_url:
    raise ValueError("GOOGLE_BASE_URL not found in environment variables. Set it before running the script.")

# Initialize the Gemini client
client = genai.Client(
    api_key=api_key,
    http_options=types.HttpOptions(
        base_url=base_url,
    ),
)

# Define prompts combining artistic styles with quality modifiers
style_quality_prompts = [
    "A beautiful landscape, photorealistic style, 4K resolution",
    "A portrait of a woman, oil painting style, high definition",
    "A futuristic cityscape, cyberpunk style, ultra-realistic, 8K resolution",
    "A serene beach scene, watercolor painting, high-quality details",
]

# Generate images for each combined prompt
for prompt in style_quality_prompts:
    response = client.models.generate_content(
        model="gemini-3.1-flash-image",
        contents=[prompt],
        config=types.GenerateContentConfig(
            response_modalities=["TEXT", "IMAGE"],
        ),
    )
    # Process and display the generated image
    for part in response.parts:
        if part.inline_data is not None:
            image = Image.open(BytesIO(part.inline_data.data))
            timestamp = int(time.time() * 1000)  # Milliseconds precision
            image_filename = f"image_{timestamp}.png"
            output_path = os.path.join("static", "images", image_filename)
            os.makedirs(os.path.dirname(output_path), exist_ok=True)
            image.save(output_path)

In this example, we define a list of prompts that combine artistic styles with quality modifiers. Each prompt specifies a style, such as "photorealistic" or "oil painting," along with quality modifiers like "4K resolution" or "high definition." The code then generates an image for each prompt using the Gemini client. The generated images are processed and saved to the specified directory.