Python in VS Code: Enhanced Code Navigation and Blazing-Fast IntelliSense (March 2026)

The March 2026 release of the Python extension for Visual Studio Code introduces two remarkable enhancements: the ability to search for symbols within installed packages and an experimental Rust-based parallel indexer that dramatically speeds up IntelliSense. These features are designed to help you navigate third-party libraries with ease and enjoy faster code completions, especially in large projects. Below, we answer common questions about these updates.

1. How can I now find functions or classes from third-party libraries without leaving VS Code?

With the new “Include Venv In Workspace Symbols” setting, Pylance can index symbols from packages installed in your active virtual environment. When you use Cmd/Ctrl+T to open Workspace Symbol search, results now include definitions from site-packages. This means you can quickly locate a function’s source code even if it resides in an external library, eliminating the need to open separate documentation or a file browser. For libraries without py.typed, only symbols exported via __init__.py or __all__ are included, keeping the results focused. This feature is opt-in to avoid performance overhead, and you can control the depth of indexing per package using the Package Index Depths setting.

2. How do I enable symbol search for installed packages?

To activate this feature, open your VS Code settings (Cmd+, on macOS or Ctrl+, on Windows/Linux). Search for “Include Venv In Workspace Symbols” and check the box under Python › Analysis. Once enabled, any Workspace Symbol search (Cmd/Ctrl+T) will include symbols from the packages in your active virtual environment. You can fine‑tune the indexing depth for individual packages by searching for Python › Analysis: Package Index Depths in settings. This allows you to decide how deeply Pylance navigates into sub‑modules, balancing search richness with performance. Remember that this setting is off by default to keep the editor responsive for everyone.

3. What is the Rust‑based parallel indexer and why is it experimental?

The Rust‑based parallel indexer is a rewritten version of Pylance’s indexing engine, which powers completions, auto‑imports, and workspace symbol search. It runs out‑of‑process and leverages parallel processing to significantly improve performance. In internal testing, it is on average 10× faster on large Python projects, resulting in quicker completions after opening a workspace and a more responsive IntelliSense experience. It is marked as experimental because the team wants to validate the gains across diverse project setups and environments before making it the default. Your feedback during this trial period is crucial for ensuring reliability and performance improvements.

4. How much faster can I expect IntelliSense to be with the parallel indexer?

Based on extensive testing, the new indexer delivers on average a 10× speed improvement on large Python projects. This means that after you open a workspace, completions and auto‑imports appear much sooner, and workspace symbol searches are snappier. For small projects you might notice little difference, but in codebases with many files and dependencies, the boost is dramatic. The exact gain depends on your project size, number of installed packages, and hardware. The feature is designed to be most impactful in complex environments, where every millisecond saved adds up to a smoother development experience.

5. How do I enable the experimental parallel indexer?

To try the Rust‑based parallel indexer, go to your settings (Cmd+, or Ctrl+,). Search for “Parallel Indexing” and check Enable Parallel Indexing (Experimental) under Python › Analysis. Alternatively, you can add the following line to your settings.json:

"python.analysis.enableParallelIndexing": true

After enabling it, reload VS Code (Cmd/Ctrl+Shift+P → Reload Window) to ensure the new indexer starts cleanly. This setting has the most effect on larger projects; small ones may see negligible change. We encourage you to test it and report any issues or improvements.

6. When I search symbols in packages, which packages are actually indexed?

The search covers packages installed in your current virtual environment’s site-packages. Only symbols from those packages that are explicitly exported through __init__.py or the __all__ variable are included, unless the package provides a type stub (py.typed). This ensures that results remain relevant and not cluttered with internal module details. You can further control indexing depth per package using the Package Index Depths setting, which lets you specify how deep Pylance should scan into sub‑modules. This targeted approach balances comprehensive code exploration with performance, making the feature useful without degrading the editor’s speed.

7. Are there any performance concerns with these new features?

Yes, indexing installed packages can impact performance, which is why symbol search for venv packages is opt‑in. By default it’s disabled, so users who don’t need it aren’t affected. Those who enable it can use the Package Index Depths setting to limit how deep indexing goes, preventing unnecessary load. The parallel indexer is also experimental; while it aims to improve performance, it may exhibit unexpected behavior in some environments. The team will collect feedback to refine it before promoting it to default. Both features are designed with performance in mind, but they give you control over the trade‑off between search richness and editor responsiveness.