Using RISE and reveal.js slides with Jupyter Notebook 6, Notebook 7, and JupyterLab

Last updated: 2025-11-16

This guide explains how to:

  • Install and use RISE in classic Jupyter Notebook 6.
  • Use RISE in a Notebook 7 environment via nbclassic.
  • Use jupyterlab-rise inside JupyterLab.
  • Understand and troubleshoot the separate reveal.js slides export based on nbconvert.

RISE is a Jupyter notebook extension that turns a live notebook into a reveal.js based slideshow right inside the browser, with code cells that you can still execute during the talk.
The reveal.js exporter from nbconvert instead generates a static HTML slide deck from a notebook.

1. Check what you have installed

In the environment where you plan to present, run:

jupyter notebook --version
jupyter lab --version  # optional
python -m pip show rise jupyterlab-rise nbclassic nbconvert

Typical cases:

  • Notebook 6.x: classic front end, compatible with rise directly.
  • Notebook 7.x: new front end, old nbextensions do not work. You need nbclassic if you want classic extensions like rise.
  • JupyterLab: use jupyterlab-rise, not the classic rise package.

2. RISE in Jupyter Notebook 6 (classic)

If you already have Notebook 6, this is the simplest path.

2.1 Install RISE

Using pip:

pip install RISE

Using conda (conda-forge or Anaconda channel):

conda install -c conda-forge rise
# or
conda install anaconda::rise

Recent RISE versions automatically install their Javascript and CSS assets when you install the package, so you usually do not need extra nbextension commands.
The RISE docs note that the manual jupyter-nbextension install step is only required for older releases. citeturn0search9turn0search11

If you want to be explicit, you can still run:

jupyter-nbextension install rise --py --sys-prefix
jupyter-nbextension enable rise --py --sys-prefix

You can check that the extension is enabled with:

jupyter-nbextension list

Look for a line that mentions rise and enabled. citeturn0search11turn0search15

2.2 Using RISE in Notebook 6

Start classic notebook:

jupyter notebook

Then:

  1. Open your notebook.
  2. Go to View -> Cell Toolbar -> Slideshow to turn on the slideshow toolbar. You can now mark cells as slide, subslide, fragment, etc. citeturn0search0turn0search21
  3. Look at the toolbar: you should see a button like “Enter/Exit RISE Slideshow” or a projector icon. Click it to enter slideshow mode.
  4. Use keyboard shortcuts (arrow keys, space, or the reveal.js controls at the bottom right) to navigate your slides.

RISE runs entirely inside the notebook, so you can execute cells live while in slideshow mode. citeturn0search0turn0search22

3. RISE in a Notebook 7 environment

The rise extension from the main RISE repository works only with the classic notebook stack (Notebook 6). It is not compatible with the new Notebook 7 front end. citeturn0search4

If you have Notebook 7 installed, you have two options:

3.1 Use nbclassic to get the classic UI

nbclassic provides the classic Notebook 6 interface as a Jupyter Server extension, side by side with Notebook 7 and JupyterLab. citeturn0search2turn0search10turn0search19

Install it in your environment:

pip install nbclassic

This automatically enables the extension in the Jupyter Server. Then start the classic interface with:

jupyter nbclassic

From this point, the steps in section 2 apply: install rise, check jupyter-nbextension list, and use the RISE button in the classic UI.

3.2 Keep using Notebook 7 without RISE

If you prefer the Notebook 7 front end and do not need RISE inside it, you can still export reveal.js slides using nbconvert (explained below).
For live-code slide decks with RISE in a Notebook 7 setup you should either:

  • Use nbclassic + rise.
  • Or switch to JupyterLab + jupyterlab-rise.

4. jupyterlab-rise in JupyterLab

For JupyterLab you should use the jupyterlab-rise extension, which is the maintained successor of RISE for the Lab ecosystem. citeturn0search5turn0search12turn0search1turn0search16

4.1 Install jupyterlab-rise

Using pip:

pip install jupyterlab-rise

Using conda:

conda install -c conda-forge jupyterlab_rise

Then restart JupyterLab:

jupyter lab

The extension registers a server extension and a Lab front end plugin. You should see a RISE related command in the command palette and possibly a button or menu entry to start a slideshow, depending on version.

4.2 Using jupyterlab-rise

The basic workflow is similar:

  1. Create or open a notebook in JupyterLab.
  2. Mark cells as slides, subslides, etc. via the slideshow cell metadata or the RISE UI.
  3. Use the RISE command (from the command palette or toolbar) to enter the slideshow.
    Behind the scenes it still uses reveal.js for transitions and navigation.

Check the jupyterlab-rise documentation for version specific screenshots and features. citeturn0search5turn0search12

5. Exporting to reveal.js slides with nbconvert

The “Download as reveal.js slides” or “Export Notebook As -> Reveal.js” menu entry in classic Notebook comes from nbconvert, not from RISE.

  • RISE is a live slideshow extension inside the notebook.
  • The reveal.js exporter from nbconvert converts a .ipynb into an HTML slide deck.

You can use this export directly from the terminal, which is also a good way to debug internal server errors:

jupyter nbconvert MySlides.ipynb --to slides --post serve

This command converts the notebook to reveal.js slides and serves them on a local HTTP server, typically at a URL like http://127.0.0.1:8000/MySlides.slides.html. citeturn0search3turn0search20

If you want to use a specific reveal.js version hosted on a CDN:

jupyter nbconvert MySlides.ipynb --to slides     --reveal-prefix "https://cdnjs.cloudflare.com/ajax/libs/reveal.js/3.3.0"

This is useful if your environment does not have reveal.js static files installed locally. citeturn0search3turn0search17

5.1 Fixing “Internal Server Error” when clicking “Export as reveal.js”

If the Notebook menu option gives a 500 Internal Server Error, try the following:

  1. Check the terminal where you launched Jupyter
    The full Python traceback usually appears there and will say if the problem is in nbconvert, a missing template, or reveal.js paths.

  2. Upgrade nbconvert and the notebook package

    pip install --upgrade nbconvert notebook

    On some setups an old nbconvert slides template is incompatible with a newer notebook front end.

  3. Test from the command line

    Run:

    jupyter nbconvert MySlides.ipynb --to slides

    If this fails, the problem is purely in nbconvert, not RISE.

  4. Use a CDN reveal-prefix

    If the error mentions reveal.js assets that cannot be found, add --reveal-prefix as shown above so the generated HTML uses hosted reveal.js instead of a local install. citeturn0search3turn0search17

  5. Check that the environment is consistent

    Sometimes Jupyter is launched from a different Python environment than the one where you installed RISE and nbconvert. Run:

    which jupyter
jupyter --paths
python -m pip show rise nbconvert

    to make sure everything lives in the same environment.

6. Quick checklist

If you are on Notebook 6 and want RISE inside the notebook:

  • pip install RISE
  • Optionally jupyter-nbextension install rise --py --sys-prefix and enable it. citeturn0search9turn0search11
  • Start with jupyter notebook.
  • Turn on slideshow toolbar and click the RISE button.

If you are on Notebook 7 and want RISE:

  • pip install nbclassic
  • pip install RISE
  • Start with jupyter nbclassic.
  • Follow the Notebook 6 steps.

If you are on JupyterLab and want RISE style slides:

  • pip install jupyterlab-rise or conda install -c conda-forge jupyterlab_rise
  • Restart JupyterLab.
  • Use the jupyterlab-rise commands to start a slideshow.

If you just want HTML reveal.js slides for your website:

  • Use jupyter nbconvert MySlides.ipynb --to slides
  • Serve or host the generated .slides.html file.
  • Use --reveal-prefix if you prefer a CDN hosted reveal.js.

With this setup you can keep classic RISE behavior for live tutorials and also export static reveal.js decks for your technical blog or conference pages.