The fastest way to start Jupyter Notebook

Open your terminal or command prompt, navigate to the folder where you want to work, and type jupyter notebook. Press Enter. Your default web browser will open to a new Jupyter Notebook interface, usually at http://localhost:8888. That's the complete process — Jupyter runs as a local server on your computer, and your browser is the interface you interact with.

The reason you use the terminal instead of clicking an icon is that Jupyter needs to know which folder to start in. When you launch it from the terminal, it uses your current working directory as the root. Any files you create or open will be relative to that location.

Key Takeaways

  • Type jupyter notebook in your terminal from the folder where you want to work, then press Enter to start the server.
  • Your browser will open automatically to http://localhost:8888, where you can create and edit notebooks.
  • If Jupyter is not installed, use pip install jupyter or conda install jupyter depending on your Python setup.
  • Stop the server by pressing Ctrl+C in the terminal window — closing the browser tab alone will not shut it down.
  • You can specify a different port or disable browser auto-launch using flags like --port=8889 or --no-browser.

Installing Jupyter if you don't have it yet

Before you can run jupyter notebook, you need Jupyter installed on your system. The most common way is through pip, Python's package manager. Open your terminal and type pip install jupyter, then press Enter. This downloads and installs Jupyter and its dependencies.

If you use Anaconda or Miniconda instead of pip, type conda install jupyter instead. Both methods work equally well — use whichever package manager you already have set up for Python on your machine.

To check whether Jupyter is already installed, type jupyter --version in your terminal. If you see a version number, you're ready to go. If you see a "command not found" error, run the install command above.

Navigating to the right folder before launching

The folder you're in when you type jupyter notebook becomes the root directory for that Jupyter session. If you want to work with files in a specific project folder, navigate there first using cd (change directory). For example, cd ~/Documents/my_project moves you into that folder.

Once you're in the correct folder, type jupyter notebook and press Enter. The Jupyter interface will open showing the contents of that folder. You can create new notebooks there, and they'll be saved in that same location.

If you launch Jupyter from your home directory or desktop and then try to navigate deep into your file system through the browser interface, it works but is slower and more error-prone than just opening the terminal in the right folder to begin with.

Understanding what happens after you press Enter

When you run jupyter notebook, several things happen at once. First, Jupyter starts a local server on your computer — you'll see output in the terminal showing the server address and a token (a long string of characters used for security). Second, your default web browser opens automatically and connects to that server.

The terminal window stays open and running. It displays log messages as you work — when you create a cell, run code, or save a notebook, you'll see entries appear in the terminal. This is normal. Do not close the terminal window unless you want to stop Jupyter entirely.

If your browser doesn't open automatically, look at the terminal output for a line that says something like http://localhost:8888/?token=abc123.... Copy that entire URL and paste it into your browser's address bar manually.

Stopping Jupyter and closing the server

To shut down Jupyter, go back to your terminal window and press Ctrl+C (or Cmd+C on Mac). You'll see a prompt asking "Shutdown this notebook server (y/[n])?" — type y and press Enter. The server stops, and you can close the terminal.

Simply closing the browser tab or window does not stop Jupyter. The server keeps running in the background, which means the port stays occupied. If you try to start Jupyter again without stopping the first instance, you'll get an error about the port already being in use.

If you accidentally closed the terminal without shutting down Jupyter properly, you can either restart your computer or use a command like lsof -i :8888 (on Mac or Linux) to find and kill the process. On Windows, use netstat -ano | findstr :8888 to locate it.

Using flags to customize how Jupyter starts

You can add flags to the jupyter notebook command to change its behavior. The most common ones are:

  • --no-browser: Starts Jupyter without opening your browser automatically. Useful if you want to open it manually or access it from a different machine.
  • --port=8889: Runs Jupyter on a different port instead of the default 8888. Use this if port 8888 is already in use.
  • --ip=0.0.0.0: Makes Jupyter accessible from other computers on your network (use with caution — this has security implications).
  • --NotebookApp.token='': Disables the security token (not recommended for shared networks).

For example, to start Jupyter on port 8889 without opening the browser, type jupyter notebook --port=8889 --no-browser. Then manually navigate to http://localhost:8889 in your browser.

Troubleshooting common startup problems

If you type jupyter notebook and see "command not found" or "not recognized", Jupyter is not installed or not in your system's PATH. Run pip install jupyter first, or check that you're using the correct Python environment if you have multiple versions installed.

If the browser opens but shows "Unable to connect" or a blank page, the server may not have started properly. Check the terminal for error messages. Common causes include port 8888 already being in use (try --port=8889) or a firewall blocking the connection.

If Jupyter starts but you can't see your files, you may have launched it from the wrong folder. Close Jupyter with Ctrl+C, navigate to the correct directory, and start again. The folder structure you see in the Jupyter browser is always relative to where you ran the command.

Frequently Asked Questions

Can I open Jupyter Notebook without using the terminal?

Some Python distributions like Anaconda include a graphical launcher that can start Jupyter without typing a command. However, you lose control over which folder Jupyter starts in, which often causes confusion when trying to find your files. Using the terminal gives you more control and is the standard approach.

What's the difference between jupyter notebook and jupyter lab?

JupyterLab is a newer interface with more features and a better file browser. If you have it installed, you can type jupyter lab instead of jupyter notebook to launch it. Both run the same notebooks — the difference is only in the interface you use to edit them.

Why does my browser show a security warning about the token?

Jupyter uses a token (visible in the terminal) as a basic security measure so only you can access your local server. This is normal and expected. You don't need to do anything — the token is automatically included in the URL when your browser opens.

Can I run Jupyter on a remote server or another computer?

Yes, but it requires additional setup. You would typically use SSH tunneling to securely connect to a remote Jupyter server. This is more advanced — start by getting comfortable with local Jupyter first, then explore remote access once you understand how the local version works.

What happens to my notebooks if I close the browser tab?

Your notebooks are saved to disk, so closing the browser tab does not delete them. However, any code you were running in cells will stop executing. The next time you open Jupyter and navigate to that notebook, your code and output will still be there.