Managing Python Environments in VS Code: Your Questions Answered
The Python Environments extension for VS Code finally unifies how you manage interpreters, environments, and packages across tools like venv, conda, pyenv, poetry, and pipenv. After a year in preview and community feedback, it's now generally available. This Q&A covers the key features—automatic discovery, faster creation with uv, project mapping, and more.
- What does the extension do and why is it needed?
- How does environment discovery work automatically?
- Can I customize which directories the extension searches?
- How does uv make environment creation faster?
- What's the difference between Quick Create and Custom Create?
- How do Python Projects handle monorepo environments?
- Will my current workflow break? Do I need to opt in?
What does the extension do and why is it needed?
Historically, Python developers in VS Code had to juggle multiple tools—venv, conda, pyenv, poetry, pipenv—each with its own commands and quirks. The Python Environments extension replaces that fragmentation with a single, consistent interface. It automatically discovers environments created by any of those tools, lets you create new ones (using your preferred manager), switch between them, and install packages—all from one place. The extension builds on the proven PET (Python Environment Tool) scanner, which already powers environment discovery in the regular Python extension, but now wraps it in a dedicated UI. For most users, everything works out of the box: open a Python file, and your environments are detected. There's no manual setup, and the extension integrates seamlessly with the existing Python extension.
How does environment discovery work automatically?
Discovery is powered by PET (Python Environment Tool), a fast Rust-based scanner that locates environments reliably across operating systems. PET checks your system’s PATH variable, known installation directories (like ~/.virtualenvs or conda envs folders), and any additional paths you configure. It recognizes environments from all major managers: venv, conda, pyenv, poetry, pipenv, and system-level Python installs. The same engine has been used in the regular Python extension for some time, so it's battle-tested. With the environments extension, the results are presented in a unified view, letting you instantly see all your environments regardless of which tool created them.
Can I customize which directories the extension searches?
Yes. For most users, the default search covers standard locations, so no configuration is needed. But if you store environments in non-standard folders, you can set workspace-level search paths using glob patterns. For example, you can point the extension to a shared envs folder inside your project. You can also configure global search paths for directories outside your workspace, such as a team-wide environment cache. This flexibility ensures that environments are found even in complex setups without interfering with your normal workflow.
How does uv make environment creation faster?
If you have uv installed on your system, the Python Environments extension uses it automatically when creating venv environments and installing packages. uv is a blazing-fast Rust-based tool that can be 10–100× faster than pip, especially for large dependency trees. The setting python-envs.alwaysUseUv is enabled by default, so you get the speed boost without any extra steps. This makes setting up a new environment feel nearly instantaneous, even when your project has heavy requirements files. If you prefer not to use uv, you can disable that setting and fall back to the standard tools.
What's the difference between Quick Create and Custom Create?
Quick Create (the + button in the Environment Manager view) is designed for speed: it builds an environment using your default manager (either venv or conda), picks the latest Python version, and automatically installs any dependencies found in requirements.txt or pyproject.toml. You get a working environment in seconds with one click.
Custom Create, invoked via the command palette (Python: Create Environment), gives you full control. You choose the environment manager (venv or conda), the exact Python version, the environment name, and which dependency files to install. For other managers like pyenv, poetry, or pipenv, the extension currently only discovers environments you create using their native CLI tools. Custom Create is ideal when you need a specific configuration or want to replicate an existing environment.
How do Python Projects handle monorepo environments?
The extension introduces the concept of Python Projects, which let you map environments to specific folders or files. This is a game-changer for monorepos (single repositories containing multiple independent projects). Instead of one environment for the whole workspace, you can assign different environments to different subfolders—for example, a Python 3.11 environment for one service and a Python 3.10 environment for another. When you open a file in a particular folder, VS Code automatically activates the correct environment. This mapping solves the common problem of cross-contamination and ensures each project uses its intended interpreter and packages.
Will my current workflow break? Do I need to opt in?
The extension is designed to be a transparent upgrade. Over the next few weeks, it will be rolled out to all users via the Python extension, and all environment workflows will switch to using it automatically. If you don’t want to wait, you can opt in immediately by setting python.useEnvsExtension to true in your settings. The extension works alongside the existing Python extension—there’s no conflict, and no setup required. All your previously created environments remain untouched. If you have environments in non-standard locations, you can configure custom search paths (see question 3). For most users, everything continues to work seamlessly, but with a better, unified experience.