Using Jupyter Notebook with a C kernel? That’s possible with the Minimal C Kernel for Jupyter. To make use of it, the best practice for the moment IMHO is to use Docker (as is suggested by the author himself), especially for Windows users. Instructions on how to use C kernel with Docker is on the GitHub page. While there has been a manual installation guide on the page too, there are some issues indeed.

Manual Installation

Works only on Linux and OS X. Windows is not supported yet. If you want to use this project on Windows, please use Docker.


  • gcc
  • jupyter
  • python 3
  • pip


  • (sudo) pip install jupyter-c-kernel <- beware that sudo may cause problems here
  • sudo install_c_kernel <- sudo is indispensable in some environments
  • jupyter notebook. Enjoy!


Above are the instructions from the original installation guide page. It says “Enjoy!”, well, not yet. Your code will still not be executable for now. The issue here is when you install with pip, it doesn’t pull in the resources/ folder as Docker does (so when you are using Docker, you will not be facing with this issue). The solution is very simple, you just need to download the folder from the repository, then put it into the folder where your jupyter-c-kernel is installed. For example, I am using anaconda, thus put it into /home/username/anaconda3/lib/python3.6/site-packages/. Then launch jupyter notebook, enjoy!