Python Works in Terminal but Fails in VS Code? Fix It Fast

Your Python script runs perfectly when you start it from the terminal. Then you press the Run button in VS Code and the same file fails with import errors, missing modules, or a message saying Python was not found. It feels like the editor is broken, but in almost every case VS Code is simply using a different Python than your terminal.

This is one of the most common setup problems in Python development. The terminal and the editor each make their own choice about which interpreter and environment to use. Once you learn how to see and control those choices, you can fix the mismatch in a few minutes and stop it from coming back.

Quick Answer

Your Python program works in the terminal but fails in VS Code because the two are using different Python interpreters or environments, so VS Code cannot see the packages your terminal installed. Open the Command Palette in VS Code, run Python: Select Interpreter, and choose the exact environment you use in the terminal. Then open a fresh integrated terminal and run the script again.

Why Terminal and VS Code Can Disagree

Most computers have more than one Python. There may be a system Python, one installed from python.org, one inside Anaconda, and a separate virtual environment for each project. Each of these has its own packages. A module installed in one is invisible to the others.

Your terminal usually activates your usual environment automatically, or you activate it yourself before working. VS Code, however, picks an interpreter based on its own settings, workspace configuration, and whatever it detected first. If that choice differs from the terminal even slightly, the editor runs your code in an environment where your packages were never installed, and imports fail immediately.

The classic symptom is ModuleNotFoundError for a package you are completely sure you installed. You did install it, just not into the Python that VS Code selected. Other symptoms include a different Python version printing than you expect, code that uses new syntax failing in the editor only, or the debugger starting a completely different environment than the Run button.

Step One: Find Out Which Python Each Side Is Using

Do not change settings blindly. First, measure the difference.

1. In your normal terminal, run python --version and then where python on Windows or which python on macOS and Linux, to see the full path of the working interpreter.
2. In the same terminal, run python -m pip list and confirm your required package appears in the list.
3. In VS Code, look at the Python version shown in the bottom status bar. Clicking it shows the interpreter path the editor is currently using.
4. Run your script in VS Code and add two temporary lines at the top: import sys and print(sys.executable). This prints the exact interpreter file VS Code launched.
5. Compare the two paths. If they are not identical, you have found your problem.

The sys.executable check is the most reliable test of all, because it reports the truth from inside the running program instead of relying on settings screens. The VS Code Python environments documentation explains how the editor discovers and selects interpreters, and it is the best reference when you manage several environments on one machine.

Step Two: Select the Correct Interpreter in VS Code

With the correct path known, tell VS Code to use it.

1. Open the Command Palette with Ctrl Shift P on Windows or Cmd Shift P on Mac.
2. Type Python: Select Interpreter and choose it from the list.
3. Pick the entry whose path matches the interpreter that works in your terminal. Choose the virtual environment inside your project folder if you have one.
4. If the right interpreter is not listed, choose Enter interpreter path and paste the full path you found earlier.
5. Confirm that the status bar now shows the expected version and environment name.

Selecting the interpreter at the workspace level saves the choice inside your project, so it survives restarts. For a deeper walkthrough of these settings, follow our step by step guide on selecting the correct Python interpreter in VS Code, which covers workspace settings and troubleshooting the picker itself.

Step Three: Use the Integrated Terminal Correctly

VS Code has its own terminal, and it does not always behave like your external one. When the Python extension is working properly, opening a new integrated terminal after selecting an interpreter should activate the chosen virtual environment automatically. You will see the environment name in brackets at the start of the prompt.

If the prompt shows no environment name, the activation did not happen. Close the old terminal completely with the bin icon rather than just opening a second tab, because an old terminal keeps the environment it started with. Then open a fresh one and check the prompt again. If needed, activate the environment manually with its activation command, for example .venv\Scripts\activate on Windows or source .venv/bin/activate on macOS and Linux.

Step Four: Install Packages Into the Right Environment

If the interpreter now matches but a package is still missing, install it using that interpreter explicitly. Run this in the VS Code terminal:

python -m pip install package_name

Using python -m pip guarantees the package lands in the environment belonging to the python command on the first line, not some other pip that happens to be on your PATH. Replace package_name with the real module distribution name you need, such as requests, pandas, or flask.

For projects with a requirements file, install everything at once with python -m pip install -r requirements.txt. Then verify with python -m pip list and try a tiny test script that only imports the package and prints its version. Small isolated tests like this separate environment problems from bugs in your real code.

Other Causes Worth Checking

Working directory differences

The terminal usually runs your script from the folder you are standing in, while VS Code may run it from the workspace root. If your code opens files with relative paths such as open("data.txt"), it may fail in one place and work in the other. Print the current folder inside your script with import os and print(os.getcwd()) in both setups, and adjust your paths or launch settings to match.

Multiple Python extensions or outdated tooling

Occasionally the Python extension itself is disabled, outdated, or confused by a stale workspace configuration. Check that the Microsoft Python extension is installed and enabled, reload the VS Code window after changing interpreters, and inspect the workspace settings file if an old interpreter path keeps coming back after you change it.

Common Mistakes and Useful Tips

• Installing packages with plain pip: Plain pip may belong to a different Python than the one running your code. Always install with python -m pip using the interpreter you verified.
• Changing the interpreter but keeping the old terminal: An open terminal keeps its original environment. Kill it and open a new one after every interpreter change.
• Creating a virtual environment but never selecting it: Creating the environment is only half the job. You must also select it in VS Code and let terminals activate it, otherwise your code still runs on the global Python.
• Guessing instead of printing sys.executable: Two lines of debug output end all arguments about which Python is running. Use them whenever behaviour differs between tools.
• Mixing Anaconda base with project environments: Installing everything into the conda base environment invites version conflicts. Keep one environment per project and select that exact environment in the editor.
• Ignoring the status bar: The interpreter shown at the bottom of VS Code is your fastest health check. Glance at it before every run, especially after opening a new project.

Conclusion

When Python works in the terminal but fails in VS Code, the code is rarely at fault. The two tools are almost always running different interpreters or environments, so one sees your installed packages and the other does not. Prove it by printing sys.executable in both places, select the matching interpreter in VS Code, open a fresh terminal so the environment activates, and install packages with python -m pip. Once both sides point at the same Python, the mysterious editor only failures disappear.

Frequently Asked Questions

Why does VS Code say Python was not found?

This happens when the interpreter VS Code is configured to use does not exist at that path, or no interpreter has been selected at all. Open Python: Select Interpreter and choose a real, installed Python. If none appear, install Python from python.org and reload the VS Code window.

Why does pip install succeed but VS Code still cannot import the package?

Because pip installed the package into a different environment than the one VS Code runs. Check the interpreter path in the status bar, then reinstall using that interpreter explicitly with python -m pip install package_name inside a fresh VS Code terminal where the correct environment is active.

Do I need a virtual environment to fix this?

Not strictly, but it is strongly recommended. A virtual environment inside your project gives both your terminal and VS Code one obvious, shared target to select. It also keeps each project’s packages separate, which prevents this class of confusion from returning with every new project.

Why does the debugger fail when the Run button works?

The debugger uses its own launch configuration, which can point to a different interpreter or working directory than a normal run. Open your launch configuration file and check the python and cwd settings, or delete the configuration and let VS Code generate a fresh one based on the selected interpreter.

How can I stop this problem from happening on a new project?

Start every project the same way: create a virtual environment, select it in VS Code immediately, install dependencies with python -m pip, and save them to a requirements file. This small routine keeps the terminal and the editor permanently in agreement.