Gemini Batch Image Editor

Batch process images using Google's Gemini 2.5 Flash Image Preview API to apply edits or generate new images.

Quick Start

1. Setup

   # Create virtual environment
   uv venv
   source .venv/bin/activate
   
   # Install dependencies
   uv pip install -r requirements.txt
   
   # Add your API key to .env file
   echo "GOOGLE_API_KEY=your-api-key-here" > .env

2. Edit Images

   # Edit single image
   python gemini_batch_image_edit.py edit image.jpg "make it vintage style"
   
   # Edit all images in directory
   python gemini_batch_image_edit.py edit ./images/ "add warm sunset lighting"

3. Generate Images

   # Generate single image
   python gemini_batch_image_edit.py generate "futuristic city at night"
   
   # Generate multiple images
   python gemini_batch_image_edit.py generate "mountain landscape" "ocean sunset" "forest path"

Features

• Batch Processing: Edit multiple images with the same prompt
• Image Generation: Create new images from text descriptions
• Flexible Input: Process single files or entire directories
• Auto-saves: Edited images saved as *_edited.png, generated as numbered files

Options

# Edit mode
python gemini_batch_image_edit.py edit [OPTIONS] INPUT PROMPT
  -o, --output DIR     Output directory (default: output)
  -m, --model MODEL    Gemini model (default: gemini-2.5-flash-image-preview)
  --extensions EXT     File extensions to process

# Generate mode  
python gemini_batch_image_edit.py generate [OPTIONS] PROMPTS...
  -o, --output DIR     Output directory (default: generated)
  -m, --model MODEL    Gemini model (default: gemini-2.5-flash-image-preview)

API Key Setup

Get your API key from Google AI Studio and add to .env:

GOOGLE_API_KEY=your-api-key-here

Notes

• Uses preview model with rate limits (2-second delay between requests)
• All outputs saved as PNG for best quality
• Generates processing_summary.json with results

License

MIT