How to use a Jupyter Notebook on Thunder Compute

If you prefer video guides, here is an overview of the following steps:

​

Connect to VSCode

Follow the instructions in our Using Thunder Compute with VSCode to set up your remote instance in VSCode.

Next, ensure you have the “Jupyter Notebook” VSCode extension.

​

Install the Jupyter extension in VSCode

Navigate to the extensions store and install the Jupyter extension within your cloud instance

​

Verify GPU availability in the notebook

Create a Jupyter Notebook, which is now connected to a Thunder Compute instance with GPU capabilities. To confirm that the GPU is accessible, run the following code in a notebook cell:

import torch
print(torch.cuda.is_available())

If the setup is correct, the output should be:

true

That’s it! You are connected to a notebook running in a Thunder Compute instance.