Quickstart

This guide walks you through installing GPU CLI, connecting your RunPod account, and running your first command on a cloud GPU.

1. Install GPU CLI

curl -fsSL https://gpu-cli.sh/install.sh | sh

This installs the gpu command to ~/.gpu-cli/bin and adds it to your PATH.

2. Login to GPU CLI

gpu login

This opens your browser to authenticate. After signing in, your CLI is connected to your GPU CLI account.

3. Connect Your RunPod Account

GPU CLI uses your RunPod API key to provision GPUs. You pay RunPod directly — GPU CLI never touches your billing.

1. Get your API key from RunPod Settings
2. Add it to GPU CLI:

gpu auth login

Follow the prompts to enter your RunPod API key.

4. Check Available GPUs

See what GPUs are available and their prices:

gpu inventory

Filter to show only available GPUs:

gpu inventory --available

5. Run Your First Command

Test that everything works:

gpu run python -c "import torch; print(f'CUDA: {torch.cuda.is_available()}, Device: {torch.cuda.get_device_name(0) if torch.cuda.is_available() else \"N/A\"}')"

GPU CLI will:

1. Provision a GPU pod (auto-selects best available)
2. Run your command
3. Stream output back to your terminal
4. Auto-stop the pod after 5 minutes of idle

6. Initialize a Project

For a real project, initialize GPU CLI in your project directory:

cd my-ml-project
gpu init

This creates a gpu.jsonc configuration file. You can customize:

• GPU type
• Output patterns to sync back
• Port forwarding
• And more

See Configuration for all options.

7. Run a Training Script

gpu run python train.py

Your code syncs to the pod, runs on the GPU, and outputs sync back automatically.

8. Check Status

See what's running:

gpu status

Or open the interactive dashboard:

gpu dashboard

Next Steps

• Configuration Reference — Customize your gpu.jsonc
• Commands Reference — All CLI commands
• Troubleshooting — Common issues and solutions