Installing Flask

Once your virtual environment is activated, install Flask:

bash

pip install Flask

Verify installation

You can quickly verify installation in Python:

bash

python -c "import flask; print(flask.__version__)"

What gets installed?

Flask depends on a few important libraries:

• Werkzeug: request/response handling, dev server
• Jinja2: templating engine
• ItsDangerous: signing cookies
• Click: CLI tooling

You don’t need to learn them all now, but it’s helpful to know what’s underneath.

Pinning versions

For learning projects, installing latest is fine.

For production apps, it’s common to pin versions in requirements.txt or use a tool like Poetry. (We’ll keep it simple in this tutorial track.)

What pip install flask actually brings

Flask is a small package that composes several others. Measured in a fresh virtual environment on CPython 3.14.4:

diagram Diagram mermaid

  flowchart TD
  F["Flask 3.1.3"] --> W["Werkzeug
WSGI, routing, the dev server"]
  F --> J["Jinja2
the template engine"]
  F --> C["click
the flask CLI"]
  F --> I["itsdangerous
signs the session cookie"]
  F --> B["blinker
signals"]
  J --> M["MarkupSafe
HTML escaping"]

measured in a clean venv	packages	site-packages
pip install flask	9	19 MB
pip install django	5	56 MB

Worth pausing on, because it contradicts the usual shorthand. Django installs fewer packages than Flask — it is one large distribution, while Flask is a thin layer over several independent libraries. Flask is smaller on disk; it is not smaller by dependency count.

The nine are Flask, Werkzeug, Jinja2, MarkupSafe, click, blinker, itsdangerous, colorama (Windows console colour) and pip itself.

Why this matters when you pin

Those are all separate projects with their own release schedules. Flask==3.1.3 alone does not pin what you got — a new Werkzeug can change behaviour under a fixed Flask version. Use pip freeze > requirements.txt, or a lock file, to record all of them.

See it move

sketch What one pip install pulls in p5.js Pause Restart

Flask is a thin layer over several independent packages, each with its own release schedule.

Check yourself

pch.quizTag pch.quizDefaultTitle

1. Measured in clean virtual environments, pip install flask brought in 9 packages and django brought in 5. What does that tell you?

   A the measurement is wrong, since Flask is the microframework B Flask is a thin layer composed of several independent packages, while Django is one large distribution C Django is missing dependencies that must be installed separately D package count is the same as install size
   pch.quizShowAnswer

   B — Flask is a thin layer composed of several independent packages, while Django is one large distribution — Flask was 19 MB against Django's 56 MB, so Flask is smaller on disk. But it is assembled from Werkzeug, Jinja2, click, itsdangerous and blinker, so it has MORE distributions, not fewer.

2. Why is Flask==3.1.3 in requirements.txt not enough to reproduce an install?

   A the version string is not valid for pip B Flask's dependencies are separate projects that release independently, so Werkzeug or Jinja2 can change under a fixed Flask C pip ignores exact pins D Flask has no dependencies to pin
   pch.quizShowAnswer

   B — Flask's dependencies are separate projects that release independently, so Werkzeug or Jinja2 can change under a fixed Flask — A new Werkzeug can alter routing or dev-server behaviour with Flask unchanged. Record every package with pip freeze or a lock file.