nbviewer is a web application that lets you enter the URL of a Jupyter Notebook file, renders that notebook as a static HTML web page, and gives you a stable link to that page which you can share with others. nbviewer also supports browsing collections of notebooks (e.g., in a GitHub repository) and rendering notebooks in other formats (e.g., slides, scripts).
Project Jupyter runs a free, public instance of nbviewer at https://nbviewer.jupyter.org. You can use it to render Jupyter Notebooks or browse notebook collection on GitHub. In either case, the notebooks must have public web URLs.
The homepage of nbviewer.jupyter.org includes some examples for you to try.
nbviewer does not execute notebooks. It only renders the inputs and outputs saved in a notebook document as a web page.
mybinder.org is a separate web service that lets you open notebooks notebooks in an executable environment, making your code immediately reproducible by anyone, anywhere. nbviewer shows an Execute on Binder icon in its navbar which
nbviewer only supports launching notebooks stored on GitHub or as Gists on Binder. Binder does support other providers directly on the mybinder.org site.
Binder tries to build a Docker image containing the notebooks and requirements
declared in a git repository. The build will fail if the repository has
environment.yaml, etc. with issues. We
suggest letting the repository owner know about the problem or submitting a
pull request to help fix it.
Binder builds a Docker image containing the notebooks in a git repository. Those notebooks may have requirements to run correctly such as libraries and data files. Binder can install these prerequisites as part of its build process, if the git repository declares them in a supported manner.
If a notebook does not run properly in its Binder environment, we suggest letting the repository owner know about the problem or submitting a pull request to help fix it.
nbviewer.jupyter.org can only render notebooks that it can access on the public Internet. If you are working on a notebook on your local machine, you need to publish that notebook somewhere with a public URL (e.g., in a GitHub repository, as a gist) in order for nbviewer.jupyter.org to render it.
Hosting your own nbviewer server opens additional avenues for rendering private notebooks. For example, an nbviewer server on your university network can render notebook files accessible via URLs on that network. Please see the README in the nbviewer repository on GitHub for instructions and options.
The URL you are visiting most likely points to a notebook that was moved or deleted. If you clicked a link on a site that lead to the 404 error page, we suggest you contact the site auownerthor to report the broken link. If a notebook author gave you the URL, we recommend asking them for an updated link.
If you notice one of the links on the nbviewer.jupyter.org, please report it as a bug in the nbviewer issue tracker.
nbviewer fetches notebooks from upstream providers (e.g., GitHub, GitHub gists, a public webserver) which host the the notebook files. You will see 4xx errors if the provider doesn't respond, the file nbviewer receives is invalid, the file is not publicly accessible, and so on.
If you believe nbviewer is incorrectly showing a 4xx error for an accessible, valid notebook URL, please file a bug in the nbviewer issue tracker.
A 5xx error or an error page from fastly may indicate that the public nbviewer.jupyter.org site is being redeployed or is down. If the problem persists for more than a few minutes, please open a bug in the nbviewer issue tracker on GitHub including the URL you are visiting and the error you receive.
nbviewer caches rendered notebooks to cut down on rendering time for popular
notebooks. The cache duration on nbviewer.jupyter.org is approximately 10
minutes. To invalidate the cache and force nbviewer to re-render a notebook
?flush_cache=true to your URL.
We originally selected notebooks that we found and liked. We are currently soliciting links to refresh the home page using a Google Form. You may also open an issue with your suggestion.
nbviewer does not store any notebooks, it only renders notebooks stored elsewhere on the web given their URLs. If you've found a notebook that you think should be removed from the web, you'll need to locate where it is hosted (e.g., on GitHub) in order to update or remove it
OVHcloud graciously hosts nbviewer.jupyter.org. Thanks to OVHcloud, we are able to provider a public nbviewer instance as a free service.
nbviewer was generously hosted by Rackspace until March, 2020.
Yes. Please do.
Yes, absolutely. Please see the README in the nbviewer repository on GitHub for instructions and options.
Please select the appropriate issue template in the nbviewer issue tracker on GitHub.
Please post your questions about using nbviewer in the Jupyter Community Forum or in the Jupyter Google Group. If you would like to propose an enhancement to nbviewer or file a bug report, please open an issue in the jupyter/nbviewer project on GitHub.