How to list installed Python packages

Listing installed Python packages is useful before upgrading dependencies, removing one package, or comparing environments that behave differently from the same project directory. A quick inventory shows what the current interpreter can already import and which distribution versions are actually present.

The python3 -m pip list command reads the installed distribution metadata for the selected interpreter and prints package names with their recorded versions. Running pip through python3 -m keeps the result attached to one concrete Python executable or virtual environment instead of whichever standalone pip command appears first in PATH.

Results vary between system installs, user-site installs, and virtual environments, so the first check is always interpreter context. On Windows, replace python3 -m pip with py -m pip when the launcher is available; inside a virtual environment created with python3 -m venv –system-site-packages, add –local to hide inherited global packages, and use pip freeze instead of treating a list output as a requirements snapshot.

Steps to list installed Python packages with pip:

Confirm which interpreter and pip installation will answer the package query.
```
$ python3 -c "import sys; print(sys.executable)"
/srv/apps/acme-api/.venv/bin/python

$ python3 -m pip --version
pip 26.0 from /srv/apps/acme-api/.venv/lib/python3.14/site-packages/pip (python 3.14)
```
Using python3 -m pip avoids mixing a different pip executable with the wrong interpreter on hosts that carry more than one Python installation.

List the installed packages in the default table format.

$ python3 -m pip list
Package            Version
------------------ ---------
certifi            2026.2.25
charset-normalizer 3.4.6
idna               3.11
markdown-it-py     4.0.0
mdurl              0.1.2
pip                26.0
Pygments           2.19.2
requests           2.33.0
rich               14.3.3
urllib3            2.6.3

Packages are listed in case-insensitive order, and editable installs add an extra column that shows the project location.

Use –user to show only user-site packages or --path /target/site-packages to inspect one installation path without changing interpreters.

Add –not-required when only top-level packages are needed instead of every dependency underneath them.
```
$ python3 -m pip list --not-required
Package  Version
-------- -------
pip      26.0
requests 2.33.0
rich     14.3.3
```
–not-required hides distributions that were installed only as dependencies, which makes ownership reviews and smaller cleanup passes easier to scan.
Add –local when a virtual environment inherits global site packages and only environment-specific packages should appear.
```
$ python3 -m pip list --local
Package  Version
-------- -------
colorama 0.4.6
pip      26.0
```
–local changes the output only inside a virtual environment with system-site access, such as one created with python3 -m venv –system-site-packages.
Select a machine-readable or requirements-style output when the package inventory needs to be compared or reused elsewhere.
```
$ python3 -m pip list --format=json
[{"name": "certifi", "version": "2026.2.25"}, {"name": "requests", "version": "2.33.0"}, {"name": "rich", "version": "14.3.3"}]

$ python3 -m pip list --format=freeze
certifi==2026.2.25
requests==2.33.0
rich==14.3.3
```
The supported pip list formats are columns, json, and freeze. The examples below show representative entries, and current pip releases do not allow –format=freeze together with –outdated.

Use pip freeze when the next step is writing a requirements snapshot, because it is the purpose-built export command and handles bootstrap packaging tools differently from pip list.

Related: How to create a requirements freeze file in Python