twat-genai 2.2.0

No project description provided

twat-genai

(work in progress)

Image generation package that leverages fal.ai's models for AI image generation. Provides a flexible command-line interface and Python API for generating images using various AI models and techniques.

Features

• Multiple AI image generation modes:
  • Text-to-image generation
  • Image-to-image transformation
  • Canny edge-guided generation
  • Depth-guided generation
• Support for LoRA (Low-Rank Adaptation) models with a built-in library of style presets
• Flexible prompt expansion with alternatives using brace syntax
• Concurrent image generation for multiple prompts
• Comprehensive metadata storage for generated images
• Modern Python packaging with PEP 621 compliance
• Type hints and runtime type checking
• Comprehensive test suite and documentation
• CI/CD ready configuration

Installation

pip install twat-genai

Usage

Command Line Interface

## Basic text-to-image generation
python -m twat_genai "a beautiful sunset" --output_dir images
## Using a specific style from the LoRA library
python -m twat_genai "a beautiful sunset" --lora "shou_xin"
## Image-to-image transformation
python -m twat_genai "enhance this photo" --model image --input_image input.jpg
## Multiple prompts with alternatives
python -m twat_genai "a {red; blue; green} house with {white; black} windows"

Python API

import twat_genai
from twat_genai.main import async_main, ModelTypes

## Generate images asynchronously
results = await async_main(
prompts="a beautiful sunset",
output_dir="generated_images",
model=ModelTypes.TEXT,
lora="shou_xin",
image_size="SQ"
)   

Key Features in Detail

Prompt Expansion

The tool supports flexible prompt expansion using brace syntax:

• "a {red; blue} house" generates two images: "a red house" and "a blue house"
• Nested alternatives are supported
• Semicolons separate alternatives

LoRA Styles

Built-in library of LoRA styles for different artistic effects:

• Gesture drawing
• Sketch and smudge effects
• 2-color illustrations
• Pencil sketches
• Tarot card style
• And more...

Image Generation Modes

• Text-to-Image: Generate images from text descriptions
• Image-to-Image: Transform existing images
• Canny Edge: Use edge detection to guide generation
• Depth-Guided: Use depth information for generation

Output Management

• Automatic file naming with customizable prefixes/suffixes
• Metadata storage in JSON format
• Various image size options (square, landscape, portrait)
• Support for custom dimensions

Development

This project uses Hatch for development workflow management.

Setup Development Environment

## Install hatch if you haven't already
pip install hatch
## Create and activate development environment
hatch shell
## Run tests
hatch run test
## Run tests with coverage
hatch run test-cov
## Run linting
hatch run lint
## Format code
hatch run format

License

MIT License
.