UV: The Fast Python Package Manager

Python UV Rust Package Manager DevTools

Why UV?

UV is very fast as compared to pip. It has been written in Rust, a systems programming language known for performance and safety. Where pip takes seconds (or minutes) to resolve dependencies, UV does it in milliseconds.

pip

Python-based

Rust-based

UV resolves and installs packages 10-100x faster than pip

💡 What UV Replaces

UV replaces multiple tools in your Python workflow: pip (installing packages), virtualenv/venv (creating environments), pyenv (managing Python versions), and pip-tools (dependency resolution). One tool. Everything.

Installation

Choose your operating system:

Install UV on Windows

Using PowerShell (recommended):

powershell

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Or using pip (if you already have Python):

cmd

pip install uv

Using WinGet:

powershell

winget install --id=astral-sh.uv -e

Install UV on macOS

Using Homebrew (recommended):

bash

brew install uv

Using the install script:

bash

curl -LsSf https://astral.sh/uv/install.sh | sh

Or using pip:

bash

pip install uv

Install UV on Linux

Using the install script (works on most distros):

bash

curl -LsSf https://astral.sh/uv/install.sh | sh

For specific distros:

bash

# Ubuntu/Debian (via apt - coming soon)
# For now, use the install script above

# Arch Linux
yay -S uv-bin

# Or using pip
pip install uv

Verify Installation

Check that UV is installed correctly:

bash

uv --version

1. List Available Python Versions

See all Python versions UV can install for you:

bash

uv python list

Note: The Python version name you get from this command is what you'll use in the next step to create your virtual environment.

2. Create a Virtual Environment

Create an isolated environment with a specific Python version:

bash

uv venv myenv --python 3.12

Note: The Python version name (e.g., 3.12) comes from the uv python list command above. UV will automatically download that version if you don't have it.

Activate the environment:

bash

# Windows
myenv\Scripts\activate

# macOS / Linux
source myenv/bin/activate

3. Start a New Project

Initialize a new project with UV (creates pyproject.toml):

bash

uv init my-project

This creates a modern Python project structure with pyproject.toml ready to go.

4. Install Packages

Install any library — UV handles dependencies in milliseconds:

bash

uv pip install python-dotenv

Install multiple packages at once:

bash

uv pip install langchain langchain-openai python-dotenv

5. Install from Requirements File

Install everything from your requirements.txt:

bash

uv pip install -r requirements.txt

Generate a requirements file from your current environment:

bash

uv pip freeze > requirements.txt

6. Run Python Scripts

Run any Python script with UV — it automatically uses the correct environment:

bash

uv run python my_script.py

✅ UV vs Pip: Quick Reference

pip install package → uv pip install package
pip install -r requirements.txt → uv pip install -r requirements.txt
python -m venv env → uv venv env
pyenv install 3.12 → uv python install 3.12
pip freeze → uv pip freeze

Next Steps

Use UV with your LangChain projects:

Simple LangChain LLMChain — Build a prompt → model → parser pipeline
Save & Load Config — Serialize prompts and LLM configs to JSON
RAG Pipeline — Build a document Q&A system with FAISS

🚀 Speed Up Your Python Workflow?

I help teams modernize their Python tooling with UV, Docker, and CI/CD pipelines.

Let's Talk