A GPU notebook in your browser

Open JupyterLab in your browser on a real NVIDIA GPU with PyTorch or TensorFlow already installed. Write code, train models and run experiments without installing CUDA or buying hardware, and pay only while it runs.

The GPU is yours alone for as long as you keep the instance: no queue, no session limits, and your files stay on its disk between sessions. You can also connect over SSH.

Which GPU to pick

For learning and coursework, a 12 GB card is plenty. For training real models, 24 GB, such as an RTX 3090 or RTX 4090, is the sweet spot of price and memory.

A working day in JupyterLab (Eight hours of notebooks and experiments) costs, on today’s cheapest hosts:

Ready-made software

Pick one of these when you rent and it starts with everything installed:

• PyTorch + JupyterLab — CUDA, PyTorch and JupyterLab ready — training, fine-tuning and notebooks. TensorBoard on port 6006.
• TensorFlow + JupyterLab — Official TensorFlow build with CUDA, JupyterLab and TensorBoard.
• CUDA dev box (bare) — Plain Ubuntu with CUDA + cuDNN, JupyterLab and SSH. Install whatever you like.

How it works

1. Create an account with Google or your email address. There is no ID upload: your first payment is how we know who you are.
2. Add credits in your own currency.
3. Add your SSH key, then pick the software and a GPU from the live board.
4. Wait for Running — about a minute on a warm host, a few minutes on a cold one — then open the app or connect over SSH.
5. Stop when you are done. GPU billing ends at once; only the disk is billed until you destroy the instance.

Questions

More things to build