USE CASE
Jupyter notebooks on a cloud GPU
Run JupyterLab with PyTorch and CUDA on a rented GPU, for data science and model experiments.
Billed per second from your USD balance. Stop or terminate the machine when you are done.
Recommended GPU
RTX 4090 · RTX 5090 · A100 SXM4
An RTX 4090 covers most notebooks. Pick an A100 when your models or datasets need more than 24 GB.
Template to pick
PyTorch
NVIDIA CUDA + PyTorch development environment, ready with SSH and Jupyter.
Up and running in four steps
- 1
Rent the machine
Open the catalog with RTX 4090 and the PyTorch template already selected and press Rent.
See RTX 4090 machines - 2
Open Jupyter
When the machine is running, press Open Jupyter on its page. JupyterLab opens in your browser; no SSH key needed.
- 3
Install what you need
PyTorch and CUDA are ready. Add other packages from a notebook cell.
%pip install transformers datasets accelerate - 4
Keep your work
Download your notebooks or push them to Git before you terminate the machine.
Questions
Is PyTorch already installed?
Yes. The PyTorch template comes with CUDA and PyTorch; add other packages with pip.
Can I keep my notebooks?
Stopping keeps the disk. Terminating deletes it, so download your notebooks or push them to Git first.