How to Open a .ipynb File (Without Jupyter)

Understanding this makes every method below make sense. A notebook is a plain JSON document — a list of cells, each with a type, its source, and (for code cells) any saved outputs. That's the whole format:

$ python -c "import json; d=json.load(open('report.ipynb')); print(list(d.keys()))"
['cells', 'metadata', 'nbformat', 'nbformat_minor']

# cells: 2  |  types: ['markdown', 'code']

Because it's just structured text, any tool that understands the format can render it — and you can even read it yourself (Method 6). The outputs (printed text, tables, and especially images) are stored inside that JSON as data, which is why notebooks get large.

The quickest path, with nothing to install: open the free Jupyter Notebook Viewer, drop your .ipynb in, and it renders the whole notebook — formatted markdown, syntax-highlighted code, and the saved outputs (text, tables, images, and error tracebacks) — right in your browser.

This is the right choice when you just need to *read* a notebook. If you're about to commit one to git, run it through the Output Cleaner first to strip the bulky outputs.

If you have Visual Studio Code, you already have a notebook viewer. Open the folder and click the .ipynb — VS Code renders it as a proper notebook (with the Python/Jupyter extension it also becomes fully editable and runnable). No browser, no server to start. This is the most convenient option for anyone who already lives in VS Code.

GitHub renders notebooks automatically: push a .ipynb to a repo and open it on github.com, and you'll see the fully rendered notebook, outputs and all — no download needed. For a notebook that lives at a public URL, nbviewer.org renders it the same way. Both are read-only, which is exactly what you want for sharing or reviewing.

If you have Python and actually want to *run* the notebook — not just read it — install Jupyter and open the real editor:

$ pip install jupyterlab
$ jupyter lab            # opens in your browser
# or the lighter classic UI:
$ jupyter notebook

This gives you the complete interactive environment: edit cells, execute them, and regenerate outputs. It's the heaviest option (a full install plus a running local server), so reach for it only when you need to *execute*, not merely view.

If you have Jupyter installed and want a shareable file — something a non-technical colleague can just double-click — convert the notebook to a self-contained HTML page:

$ jupyter nbconvert --to html report.ipynb
# writes report.html — opens in any browser, no Python needed

For automation — or if you just want the text out of a notebook — you don't need any package. Since it's JSON, the standard library reads it in a few lines:

import json

nb = json.load(open('report.ipynb'))
for cell in nb['cells']:
    src = ''.join(cell['source'])
    tag = 'MD' if cell['cell_type'] == 'markdown' else 'CODE'
    print(f'[{tag}] {src.splitlines()[0]}')

# [MD]   # Sales report
# [CODE] import pandas as pd

That's the same idea every viewer uses under the hood. If what you actually want is the runnable Python pulled out into a .py file, use the Jupyter → Script converter or see the full guide on converting a notebook to a script.

For the overwhelmingly common case — *"I just need to see what's in this notebook"* — the online viewer wins on speed and requires nothing. Everything else is about what you plan to do *after* opening it.