au is a fast Python version and package management tool inspired by au and conda, providing features for installing Python versions, managing virtual environments, and installing packages.

• 🚀 Fast Installation - Quick download and installation of Python versions
• 🐍 Version Management - Easy management of multiple Python versions
• 📦 Virtual Environments - Fast creation and management of virtual environments
• 📋 Package Management - Quick installation and management of Python packages
• 🏗️ Project Management - au-like project dependency management
• 🔍 Smart Resolution - Advanced dependency conflict detection and resolution
• 💾 Efficient Caching - Content-addressable cache with LRU eviction strategy
• ⚙️ Configuration Management - Flexible YAML configuration options
• 🔄 Sync Functionality - conda-like environment synchronization mechanism

# Build from source
git clone https://github.com/ForesAi/au.git
cd au
go build -o au main.go

# Or use go install
go install github.com/ForesAi/au@latest

# Install Python versions
au python install 3.11.0
au python install 3.12.0

# List installed Python versions
au python list

# Set default Python version
au python use 3.11.0

# Create virtual environment
au venv create myproject
au venv create myproject --python 3.12.0

# List virtual environments
au venv list

# Activate virtual environment
au venv activate myproject

# Remove virtual environment
au venv remove myproject

# Install packages
au pip install requests
au pip install numpy pandas

# Install from requirements file
au pip install -r requirements.txt

# List installed packages
au pip list

# Uninstall packages
au pip uninstall requests

# Export dependencies
au pip freeze > requirements.txt

au uses the ~/.au.yaml configuration file. You can customize the following settings:

default_python: "3.11.0"
python:
  install_dir: "~/.au/python"
venv:
  dir: "~/.au/venv"
pip:
  index_url: "https://pypi.org/simple/"
  extra_indexes: []
  timeout: 30
  retries: 3
cache:
  dir: "~/.au/cache"
  max_size: "1GB"
  ttl: "7d"

Python version management commands.

Subcommands:

• install <version> - Install specified Python version
• list - List installed Python versions
• use <version> - Set default Python version

Virtual environment management commands.

Subcommands:

• create <name> - Create new virtual environment
• list - List all virtual environments
• remove <name> - Remove virtual environment
• activate <name> - Show command to activate virtual environment

Package management commands.

Subcommands:

• install <packages...> - Install Python packages
• list - List installed packages
• uninstall <packages...> - Uninstall Python packages
• freeze - Output installed packages in requirements format

Options:

• -r, --requirement <file> - Install from requirements file
• --editable - Install package in editable mode
• --index-url <url> - Specify package index URL
• --upgrade - Upgrade specified packages

go build -o au main.go

go test ./...

Issues and Pull Requests are welcome!

MIT License

Feature	au	au	conda	pip
Python Version Management	✅	❌	❌	❌
Virtual Environment Management	✅	✅	✅	❌
Package Installation Speed	🚀 Fast	🚀 Very Fast	🐢 Slower	🐢 Slower
Dependency Resolution	🔍 Smart Resolution	🔍 Smart Resolution	🔍 Smart Resolution	❌ Basic
Lock File Support	✅ au.lock	✅ au.lock	❌	❌
Project Management	✅ pyproject.toml	✅ pyproject.toml	❌	❌
Multi-Environment Support	✅ dev/prod/test	✅ dev/prod	✅ environments	❌
Caching Mechanism	✅ Content-addressable+LRU	✅ Content-addressable	✅ Index Cache	❌
Concurrent Installation	✅ Supported	✅ Supported	❌	❌
Configuration File	✅ au.yaml	✅ pyproject.toml	✅ .condarc	❌
Cross-Platform Support	✅ Linux/Windows/macOS	✅ Linux/Windows/macOS	✅ Linux/Windows/macOS	✅ All Platforms
Docker Support	✅ Official Image	✅ Official Image	✅ Official Image	❌
GitHub Actions	✅ Auto Build	✅ Auto Build	✅ Auto Build	❌
Binary Distribution	✅ Multi-architecture	✅ Multi-architecture	✅ Multi-architecture	❌

Operation	au	au	conda	pip
Create Virtual Environment	~2s	~1s	~5s	~2s
Install requests Package	~3s	~2s	~8s	~5s
Install 10 Packages	~15s	~10s	~45s	~30s
Dependency Resolution (50 packages)	~5s	~3s	~20s	N/A
Cache Hit Rate	85%	90%	70%	N/A

Scenario	au	au	conda	pip
Web Development	✅ Recommended	✅ Recommended	⚠️ Available	✅ Basic
Data Science	✅ Good	⚠️ Limited	✅ Best	⚠️ Limited
Machine Learning	✅ Good	⚠️ Limited	✅ Best	⚠️ Limited
CI/CD	✅ Fast	✅ Fast	⚠️ Slower	⚠️ Slower
Enterprise Environment	✅ Flexible	✅ Simple	✅ Mature	⚠️ Basic

Download binary files for corresponding platforms from GitHub Releases:

# Linux (x86-64)
wget https://github.com/ForesAi/au/releases/latest/download/au-linux-amd64.tar.gz
tar -xzf au-linux-amd64.tar.gz
sudo mv au /usr/local/bin/

# macOS (Apple Silicon)
wget https://github.com/ForesAi/au/releases/latest/download/au-darwin-arm64.tar.gz
tar -xzf au-darwin-arm64.tar.gz
sudo mv au /usr/local/bin/

# Windows (x86-64)
wget https://github.com/ForesAi/au/releases/latest/download/au-windows-amd64.zip
unzip au-windows-amd64.zip
# Add au.exe to PATH

# Using go install
go install github.com/ForesAi/au@latest

# Using Homebrew (macOS)
brew install ForesAi/au/au

# Using Docker
docker pull ForesAi/au-tool/au
docker run --rm -it ForesAi/au-tool/au --help

git clone https://github.com/ForesAi/au.git
cd au
go build -o au main.go

• au - Inspiration source, fast package management
• conda - Environment management and dependency resolution reference
• spf13/cobra - CLI framework
• pip - Python package management foundation