Table of contents

Quick Start Guide

This guide takes you from a brand-new account to a running GPU pod you can SSH into — about 10 minutes. Each step links to a deeper guide if you want detail.

What you’ll accomplish

Sign in, get verified, set up a project and key, fund your wallet, launch a pod, and connect to it.

The path

1. Sign in

Open the portal and sign in with an email code or a social login. Your first sign-in creates your account. → Creating Your Account

2. Create a project

Create a project so your resources and costs have a home. It becomes your active context automatically. → Creating & Switching Projects

3. Add an SSH key

Generate a key in-platform (download the private key once) or upload an existing public key. → SSH Keys

4. Fund your wallet

Add funds so you can launch paid resources. Billing is per-second — you pay only for the GPU fraction and the seconds you use. → Wallet & Billing

5. Launch a pod

Console: go to Pods → Create Pod, pick a GPU (fractional or full), choose a template or image, select your SSH key, and launch. For the full walkthrough, see Creating Pods.

CLI: the fastest path from a terminal:

# Interactive: arrow-select a GPU, pick count + SSH key, confirm
podstack launch

# Or scripted, with flags
podstack gpu instances create --gpu-type <type> --gpu-count 1

See the CLI section for setup and full commands.

6. Connect

ssh root@<pod-address>
# or, with a specific key
ssh -i ~/.ssh/podstack_key root@<pod-address>

See Connecting to Pods for SSH config, port forwarding, and Windows instructions.

Verify it worked

  • The pod shows Running in the Pods list (or podstack gpu instances list).
  • You get a shell after ssh, greeted by the PodStack banner.
  • The dashboard’s Run Rate card now shows an hourly cost.

What’s next

  • Compute — pods, VMs, sandboxes, notebooks, the GPU marketplace.
  • Storage — object storage, volumes, and NFS for your data.
  • Inference — serve models on OpenAI-compatible endpoints.
  • CLI — manage everything from your terminal.