What's New in Python for Visual Studio Code: March 2026 Update

By • min read

Microsoft has rolled out the March 2026 release of the Python extension for Visual Studio Code, bringing two significant enhancements designed to streamline code navigation and boost performance. Developers can now search for symbols across installed packages directly from their workspace, and an experimental Rust-based parallel indexer promises to dramatically accelerate IntelliSense responses. Below, we break down what these updates mean for your daily coding workflow.

Working with unfamiliar libraries or large codebases often requires jumping into third-party packages to find function and class definitions. Previously, this meant leaving VS Code to consult external docs or manually browsing the site-packages folder. The March 2026 release addresses this by extending the Workspace Symbol search (triggered via Cmd/Ctrl+T) to include symbols from packages installed in your active virtual environment.

What's New in Python for Visual Studio Code: March 2026 Update
Source: devblogs.microsoft.com

How It Works

Pylance, the language server powering Python IntelliSense, now indexes symbols from packages within your virtual environment’s site-packages directory. When you open the symbol search, results will surface relevant items from those external libraries alongside your own code. For packages that do not provide a py.typed marker, only publicly exported symbols—those listed in __init__.py or __all__—are included to keep results focused.

Configuration

Because indexing every installed package can impact performance, this feature is opt-in. To enable it, navigate to VS Code settings (Cmd+, on macOS or Ctrl+, on Windows/Linux) and search for “Include Venv In Workspace Symbols”. Check the box under Python > Analysis to activate it. Additionally, you can fine-tune the depth of indexing per package using the “Package Index Depths” setting, which controls how deeply Pylance searches into sub-modules.

Benefits

This enhancement is especially valuable when onboarding into a new project or exploring unfamiliar dependencies.

Experimental Rust-Powered Parallel Indexer

Behind the scenes, Pylance relies on an indexer to provide completions, auto-imports, and symbol searches. The March 2026 release includes an experimental setting that switches this indexer to a new Rust-based parallel implementation that runs out-of-process. Early benchmarks show an average 10× speed improvement on large Python projects, meaning faster completions immediately after workspace open and a more responsive IntelliSense experience.

What's New in Python for Visual Studio Code: March 2026 Update
Source: devblogs.microsoft.com

Performance Improvements

The new indexer leverages Rust’s performance and memory safety to process files in parallel, significantly reducing the time needed to build a symbol database. In tests with sizable codebases (thousands of files), the initial indexing time dropped from several seconds to under a second in many cases. Subsequent code edits also benefit from quicker partial re-indexing.

How to Enable

Since this is an experimental feature, it must be explicitly turned on. Go to VS Code settings and search for “Parallel Indexing”, then check “Enable Parallel Indexing (Experimental)” under Python > Analysis. Alternatively, add the following to your settings.json:

"python.analysis.enableParallelIndexing": true

After enabling, reload VS Code (Cmd/Ctrl+Shift+PReload Window) to ensure the new indexer starts cleanly. Note that on very small projects, the difference may be negligible, but larger codebases will see the most benefit.

Expected Impact

Microsoft encourages users to test this feature in their own environments and provide feedback to help refine it before it becomes the default in a future release.

Conclusion

The March 2026 update brings practical improvements to code exploration and performance. The new symbol search across installed packages makes it easier to understand dependencies without leaving VS Code, while the experimental Rust-based indexer promises a snappier experience for large projects. Both features are optional, allowing developers to control when and how they adopt them. Check the full changelog for additional minor enhancements and bug fixes.

Recommended

Discover More

Breaking: Internal Search Failures Drive Users to Google — New Analysis Exposes the 'Site Search Paradox'10 Harsh Realities of Dying in Space: Separating Sci-Fi from ScienceHow to Spot and Avoid Rogue AI Browser Extensions That Steal Your DataWhen Observability Becomes Dependency: Hyrum's Law, Restartable Sequences, and the TCMalloc DilemmaSelf-Service API Migrations: The Source-Level Inliner in Go 1.26