The Python extension for Visual Studio Code has just received its March 2026 update, bringing powerful new capabilities for developers who work with large codebases or explore unfamiliar libraries. This release focuses on smarter code navigation and blazing-fast indexing performance. Whether you're a data scientist, web developer, or hobbyist, these improvements aim to make your coding experience smoother and more efficient. Let's dive into the top five features you need to know about.
1. A Fresh Overview of the March 2026 Update
This month's release marks a significant step forward for the Python and Pylance extensions. The most notable additions are the ability to search for symbols in installed packages and an experimental Rust-based parallel indexer. Both features are designed to reduce friction when working with dependencies and large projects. The update is opt-in, meaning you can choose to enable these enhancements without affecting your current workflow. The development team has focused on performance and reliability, ensuring that the default experience remains fast and stable. If you're using a virtual environment, you'll now have richer access to third-party code directly within your editor. This release also includes several bug fixes and minor improvements, which you can explore in the official changelogs. Overall, it's a compelling update that empowers you to navigate code more intelligently.
2. Installed Package Symbols Now in Workspace Search
One of the most practical improvements is the ability to include symbols from installed packages when using Workspace Symbol search (Cmd/Ctrl+T). Previously, this search only covered files within your workspace, leaving you to manually locate definitions in site-packages. Now, by enabling the Python › Analysis: Include Venv In Workspace Symbols setting, Pylance will index symbols from packages in your active virtual environment. This means you can quickly jump to a function or class definition from a library like NumPy or Flask without leaving VS Code. For libraries without py.typed markers, only symbols exported via __init__.py or __all__ are included, keeping the results relevant and focused. This feature is opt-in to avoid performance impacts on all users. To enable it, open Settings, search for "Include Venv In Workspace Symbols", and check the box.
3. Fine-Tuning Package Indexing Depth
While the new symbol search is powerful, indexing every package can be resource-intensive. That's why the March 2026 update introduces a complementary setting: Python › Analysis: Package Index Depths. This allows you to control how deeply Pylance searches into sub-modules of each package. For example, you might want full searchability for a commonly used library like requests, but only shallow indexing for a package with many internal modules. By adjusting this per-package depth, you can strike a balance between comprehensive code exploration and performance. This setting is especially useful in monorepos or projects with many dependencies. To configure it, go to your settings and look for "Package Index Depths". You can define a list of package names and their desired search depth (e.g., 2 for sub-modules). This granular control ensures that the symbol search remains responsive even as your environment grows.
4. Experimental Rust-Based Parallel Indexer Arrives
Under the hood, Pylance's indexing engine has been given a major overhaul. The experimental Python › Analysis: Enable Parallel Indexing setting switches the indexer to a new Rust-based implementation that runs out-of-process. In internal tests, this indexer is on average 10× faster on large Python projects. This means quicker completions, faster auto-imports, and a more responsive IntelliSense experience immediately after opening a workspace. The indexer runs in parallel, leveraging multiple CPU cores and avoiding blocking the main editor thread. Because this is experimental, it has intentionally not been made the default. The team wants to validate performance gains and reliability across diverse project setups. Small projects may not see a noticeable difference, but for large codebases, the speed boost can be transformative. To enable it, search for "Parallel Indexing" in Settings, check the box, then reload VS Code (Ctrl+Shift+P → Reload Window).
5. How to Enable and Test the Indexing Features
Both new features are opt-in, so you'll need to explicitly activate them. For symbol search in packages, open Settings (Cmd+, / Ctrl+,), search for "Include Venv In Workspace Symbols", and check the box. To fine-tune indexing depth, look for "Package Index Depths" and configure per-package levels. For the parallel indexer, search for "Parallel Indexing" and enable the experimental setting. You can also add "python.analysis.enableParallelIndexing": true directly to your settings.json. After making changes, especially for the parallel indexer, reload VS Code to ensure the new engine starts cleanly. We encourage you to test these features on your own projects. The parallel indexer has the most impact on larger codebases. If you encounter any issues or notice improvements, please provide feedback through the Python extension's GitHub repository. Your input will help shape the default behavior in future releases.
This March 2026 update brings meaningful improvements to code navigation and performance. By enabling these opt-in features, you can explore third-party libraries with ease and enjoy faster indexing thanks to Rust-based parallelism. Whether you're managing a complex monorepo or just curious about a new package, these tools make Visual Studio Code an even more powerful Python editor. Try them out and share your experience with the community!