Skip to main content

​
Pixi: Package Management Made Easy

Pixi is a cross-platform, multi-language package manager and workflow tool built on the foundation of the conda ecosystem. It provides developers with an exceptional experience similar to popular package managers like cargo or npm, but for any language.
Pixi is ready for production! The file format is designed to be compatible with previous versions, ensuring you can rely on Pixi with peace of mind.

​
Why Pixi?

Reproducible Environments

Built-in lock files ensure your environments are isolated and easily recreated across machines. No more “works on my machine” problems.

Multi-Language Support

Supports Python, C++, R, Rust, Node.js, and more using Conda packages. Over 30,000 packages available on conda-forge.

Cross-Platform

Works seamlessly on Linux, Windows, macOS (including Apple Silicon), and more platforms.

Built-In Task Runner

Define and run cross-platform tasks effortlessly. Chain commands, set dependencies, and manage complex workflows.

Disk Efficient

Environments share files through hard links or reflinks, so packages are stored only once on your system.

PyPI Integration

Full support for PyPI packages through uv, seamlessly integrated with conda packages.

​
Key Features

​
Lock Files for Reproducibility

Pixi always creates a pixi.lock file containing exact versions of all dependencies and their dependencies. This ensures reproducible environments across different machines and over time.

​
Multiple Environments in One Workspace

Compose multiple environments in a single manifest file. Perfect for managing development, testing, and production environments together.

​
Global Tool Installation

Install tools globally, safely isolated from each other. Pixi can replace apt, homebrew, or winget for managing your development tools: 
pixi global install gh nvim ipython btop ripgrep

​
Simple, Cargo-Like CLI

Pixi provides a clean and intuitive command-line interface that feels familiar to developers: 
pixi init my-project
pixi add python numpy pytest
pixi run test

​
How Does Pixi Compare?

| Core Features | Pixi | Conda | Pip | Poetry | uv | |-----------------------------|------|-------|-----|--------|----|--- | Installs Python | ✅ | ✅ | ❌ | ❌ | ✅ | | Supports Multiple Languages | ✅ | ✅ | ❌ | ❌ | ❌ | | Lockfiles | ✅ | ❌ | ❌ | ✅ | ✅ | | Task Runner | ✅ | ❌ | ❌ | ❌ | ❌ | | Workspace Management | ✅ | ❌ | ❌ | ✅ | ✅ |

​
Available Software

Pixi defaults to the conda-forge repository, which contains over 30,000 packages including:
  • Python: python, scikit-learn, pytorch, numpy, pandas
  • C/C++: clang, boost, opencv, cmake, ninja
  • Java: openjdk, gradle, maven
  • Rust: rust, cargo-edit, cargo-insta
  • Node.js: nodejs, bun, pnpm, eslint
  • CLI Tools: git, gh, ripgrep, make, jq
Browse thousands more on prefix.dev or host your own channels.

​
Built with Rust

Pixi is entirely written in Rust and built on top of the rattler library, ensuring exceptional performance and reliability.

​
Get Started

Installation

Install Pixi on Linux, macOS, or Windows in seconds

Quick Start

Create your first workspace and run your first task

Basic Usage

Learn the essential commands and workflows

​
What’s Next?

Some exciting features planned for upcoming releases:
  • Build and publish your project as a Conda package
  • Support for dependencies from source
  • More powerful global installation of packages for deterministic setups across machines

​
Community

Discord

Join our active community for help and discussion

GitHub

Star the project and contribute to development