Python NumPy Not Found: How to Fix No Module Named NumPy

Your script dies on line one with ModuleNotFoundError: No module named 'numpy'. NumPy usually isn’t missing from your computer. It’s missing from the Python that’s running your code.

Quick answer: a Python NumPy not found error means the interpreter running your script has no NumPy package. Install it into that same interpreter with python -m pip, then rerun the script with the same python command:

python -m pip install numpy

In this tutorial, we’ll fix Python NumPy not found errors the way our team does it: reproduce the error, find which Python can’t see NumPy, install it there and confirm the import. We ran every step on Windows with Python 3.14.8 and NumPy 2.5.4.

What You Need Before You Start Fixing NumPy Not Found

• A terminal: Command Prompt or PowerShell on Windows, Terminal on macOS or Linux.
• Python 3.12 or newer. NumPy 2.5.4 declares Requires-Python >=3.12.
• An internet connection, so pip can download NumPy.

Our running sample is sales.py, a four-week sales total. If you’re unsure how to start a script from the terminal, see how to run a Python script.

Step 1: Reproduce the NumPy not found error

We always run the failing script first, so we know exactly what we’re fixing. Here’s sales.py in a fresh virtual environment (a private Python folder with its own packages) that doesn’t have NumPy yet:

import numpy as np

sales = np.array([1249.99, 879.50, 1530.25, 990.00])   # weekly sales, USD
print(f"Total: ${sales.sum():,.2f}")
print(f"Average week: ${sales.mean():,.2f}")

Output:

Traceback (most recent call last):
  File "C:\pyguides\sales.py", line 1, in <module>
    import numpy as np
ModuleNotFoundError: No module named 'numpy'

The last line names the missing module, and it fails on the import itself. The code is fine; the environment isn’t.

Step 2: Check which Python can’t find NumPy

Most machines have more than one Python, and each has its own packages. We print the running interpreter, then ask its pip about NumPy. More on juggling installs is in installing multiple Python versions.

import sys

print("Python", sys.version.split()[0])
print("Running from:", sys.executable)

Output:

Python 3.14.8
Running from: C:\pyguides\npdemo\Scripts\python.exe
WARNING: Package(s) not found: numpy
Name: numpy
Version: 2.5.3
Summary: Fundamental package for array computing in Python
Home-page: https://numpy.org
Author: Travis E. Oliphant et al.
Author-email: 
License: 
Location: C:\pyguides\venv\Lib\site-packages
Requires: 
Required-by: contourpy, formulaic, matplotlib, opencv-python, pandas, patsy, scikit-learn, scipy, statsmodels

That’s the classic trap: NumPy was installed, but into C:\pyguides\venv, and the script runs from C:\pyguides\npdemo. A plain pip command can point at either one.

Step 3: Install NumPy with python -m pip

python -m pip runs pip through the exact interpreter you call, so the package lands where your script looks. We install, then rerun the same file:

python -m pip install numpy
python sales.py

Output:

Collecting numpy
  Using cached numpy-2.5.4-cp314-cp314-win_amd64.whl.metadata (6.6 kB)
Using cached numpy-2.5.4-cp314-cp314-win_amd64.whl (12.7 MB)
Installing collected packages: numpy
Successfully installed numpy-2.5.4
Total: $4,649.74
Average week: $1,162.43

The sample now prints Total: $4,649.74 and Average week: $1,162.43. Need an older release for a legacy project? See installing a specific version with pip.

Note: on Windows, py -m pip install numpy uses the Python launcher’s default Python, and on macOS or Linux use python3 -m pip. Run the script with that same command.

Step 4: Confirm NumPy imports in the right Python

We finish with a two-line check that prints the versions and where NumPy was loaded from:

import sys
import numpy

print("Python", sys.version.split()[0], "| NumPy", numpy.__version__)
print("Loaded from:", numpy.__file__)

Output:

Python 3.14.8 | NumPy 2.5.4
Loaded from: C:\pyguides\npdemo\Lib\site-packages\numpy\__init__.py

The path points inside npdemo, so it’s the copy we just installed. Our guide to checking your Python version shows more ways to read the interpreter version.

How do you fix NumPy not found in VS Code, Jupyter or Anaconda?

It’s the same cause: the editor or kernel runs a different Python than your terminal. In VS Code, run Python: Select Interpreter from the Command Palette and pick the environment that has NumPy.

In a Jupyter notebook, run %pip install numpy in a cell, since it installs into the kernel’s Python, then restart the kernel. In Anaconda, use conda install numpy in the active environment. ModuleNotFoundError after pip install covers the editor cases in more depth.

NumPy Not Found Cheat Sheet

Here’s the quick map we use for this error:

What you see	Likely cause	Fix
No module named 'numpy'	NumPy not in the running Python	python -m pip install numpy
Works in terminal, fails in VS Code	Editor uses another interpreter	Python: Select Interpreter
Fails in Jupyter only	Kernel uses another Python	%pip install numpy, restart kernel
No module named 'Numpy'	Wrong case in the import	import numpy
has no attribute 'array'	Your own file named numpy.py	Rename the file

Common Mistakes to Avoid with NumPy Imports

Two mistakes look like a missing install but aren’t. First, module names are case-sensitive, even on Windows:

import Numpy as np

Output:

Traceback (most recent call last):
  File "C:\pyguides\report.py", line 1, in <module>
    import Numpy as np
ModuleNotFoundError: No module named 'Numpy'

Second, a file of your own named numpy.py sits first on the import path, so Python imports it instead of the real library:

import numpy as np

print(np.array([19.99, 5.49]))

Output:

Traceback (most recent call last):
  File "C:\pyguides\npshadow\prices.py", line 3, in <module>
    print(np.array([19.99, 5.49]))
          ^^^^^^^^
AttributeError: module 'numpy' has no attribute 'array'

Rename that file and delete any __pycache__ folder next to it. A different message, numpy.core.multiarray failed to import, is a version clash instead; see fixing numpy.core.multiarray failed to import.

Next Steps after NumPy imports

With NumPy working, I’d start with arrays you can see and print. Our tutorial on making a matrix in Python is a good next stop.

Here are a few more NumPy and pip guides we point teammates to:

• Generate random numbers with NumPy
• Convert a NumPy array to a list
• Install packages from requirements.txt
• Fix pip’s exit status 1 error

Frequently asked questions