Gemini Image Generation

Generate images using Google's Gemini 3 Pro image generation API.

Requirements

• GEMINI_API_KEY environment variable set, or passed via --api-key
• google-genai Python package installed (pip install google-genai)

Usage

Command Line

# Basic usage
python scripts/generate_image.py "a cat wearing a spacesuit"

# With options
python scripts/generate_image.py "a futuristic city" \
    --output cityscape \
    --aspect-ratio 16:9 \
    --size 2K

From Python

from scripts.generate_image import generate_image

files = generate_image(
    prompt="a serene mountain landscape at sunset",
    output_name="mountain",
    aspect_ratio="16:9",
    size="2K"
)

Parameters

• prompt (required): Text description of the image to generate
• output_name: Base filename for output (default: "generated_image")
• aspect_ratio: Image proportions
  • 1:1 - Square (default)
  • 16:9 - Widescreen
  • 4:3 - Standard
  • 3:4 - Portrait
  • 9:16 - Mobile vertical
• size: Resolution
  • 1K - 1024px (default)
  • 2K - Higher resolution
  • 4K - Maximum resolution

Output

Generated images are saved to the current directory with auto-detected extensions (usually .png or .jpg). Multiple images may be generated for a single prompt.