Skip to main content

Unable to Detect Venv as Kernel in VS Code

Updated Jul 11, 2026 ·

Overview​

Use this when a .venv exists for a Python project, but VS Code does not show it as a Jupyter notebook kernel.

This commonly happens when:

  • VS Code is opened at the wrong folder.
  • ipykernel is missing from the project environment.
  • The notebook is attached to a global Python install instead of the project .venv.
  • VS Code is running on Windows while the project environment is inside WSL.

Open the project folder directly in VS Code.

If the project is inside a parent folder, avoid opening only the parent folder unless you intentionally want that parent folder as the workspace.

Example:

Git/
|-- project-a/
|   |-- .venv/
|   |-- pyproject.toml
|   `-- notebook.ipynb
`-- project-b/

For project-a, open the project-a/ folder directly in VS Code.

macOS​

Open the project folder:

cd ~/Git/project-a
code .

Install ipykernel in the project environment:

uv add --dev ipykernel
uv sync

Verify the environment:

uv run python --version
uv run python -m ipykernel --version

If VS Code still does not show the environment, register a named kernel:

uv run python -m ipykernel install \
  --user \
  --name project-a \
  --display-name "Python (.venv project-a)"

Reload VS Code:

  1. Open the command palette.
  2. Run Developer: Reload Window.
  3. Open the notebook again.
  4. Select Python (.venv project-a) from the kernel dropdown.

Windows with PowerShell​

You can use this when the project and .venv are created directly on Windows.

Open a Powershell terminal and navigate to the project folder:

cd C:\Git\project-a
code .

Install ipykernel in the project environment:

uv add --dev ipykernel
uv sync

Verify the environment:

uv run python --version
uv run python -m ipykernel --version

If needed, register a named kernel:

uv run python -m ipykernel install --user --name project-a --display-name "Python (.venv project-a)"

In VS Code:

  1. Open the command palette.
  2. Run Developer: Reload Window.
  3. Open the notebook again.
  4. Select the project .venv or the named kernel.

Windows with VS Code and WSL​

Use this when the project .venv is inside WSL.

Open the project from a WSL terminal:

cd ~/Git/project-a
code .

Note that this opens VS Code on a remote server inside WSL. You should see WSL: Ubuntu-22.04 in the lower-left corner of VS Code. Another way to check is to open the command palette and run Remote-WSL: New Window. If it opens a new VS Code window, you are connected to WSL.

Install ipykernel inside WSL:

uv add --dev ipykernel
uv sync

Verify the WSL environment:

uv run python --version
uv run python -m ipykernel --version

Check the actual .venv Python path:

.venv/bin/python --version

Open the notebook in VS Code and select the project .venv from the kernel dropdown.

If VS Code still does not detect the kernel, register it from WSL:

uv run python -m ipykernel install \
  --user \
  --name project-a \    ### change to a unique name if you have multiple projects
  --display-name "Python (.venv project WSL)"

Then reload VS Code and select the named kernel.

Do not mix Windows and WSL environments

If the project lives in WSL, install ipykernel from the WSL terminal.

If the project lives on Windows, install it from PowerShell. Mixing the two can make VS Code show the wrong interpreter.

If you get this when trying to select the kernel:

Install/Enable suggested extensions 
Browse marketplace for extensions

This means the Jupyter extension is not installed or enabled inside the WSL VS Code environment.

Extensions installed on Windows are separate from extensions installed in the WSL extension host, so they may appear installed locally but still need installation in WSL.

Open Extensions and search the following:

  • Python
  • Jupyter

Select the extensions and click:

Install in WSL: Ubuntu

or:

Enable in WSL: Ubuntu

Reload VS Code and open the notebook again. You should now see the project .venv in the kernel dropdown.

Quick Checks​

Check whether ipykernel is installed:

uv run python -m ipykernel --version

For a regular activated environment, this also works:

python -m ipykernel --version

For macOS, Linux, and WSL, check the project interpreter:

.venv/bin/python --version

For Windows PowerShell:

.venv\Scripts\python.exe --version

Recreate the Environment​

If the environment was created with the wrong Python version or missing packages, recreate it.

For macOS, Linux, and WSL:

rm -rf .venv
uv sync
uv add --dev ipykernel

For Windows PowerShell:

Remove-Item -Recurse -Force .venv
uv sync
uv add --dev ipykernel

Reload VS Code after recreating the environment.

Cleanup​

List registered Jupyter kernels:

jupyter kernelspec list

Remove an old named kernel:

jupyter kernelspec remove project-a

If jupyter is not available globally, run it through uv:

uv run --with jupyter jupyter kernelspec list
uv run --with jupyter jupyter kernelspec remove project-a

Remove a broken project .venv and rebuild it:

rm -rf .venv
uv sync

For Windows PowerShell:

Remove-Item -Recurse -Force .venv
uv sync
Only remove the project environment

Run cleanup commands from the project folder, and confirm that .venv is the environment you want to recreate.

References​