Skip to main content

Quickstart: Deploy Your First Pod

This page walks you through deploying your first GPU container (Pod) and connecting to it through the Web Terminal.

Prerequisites​

Step 1: Select a resource​

  1. Go to Cloud > Pods and click Create Pod.
  2. Choose a suitable GPU specification from the resource cards; you can filter by brand, model, or GPU mode. The stock indicator on each card shows real-time availability (green = in stock, yellow = limited, red = low, black = out of stock).
  3. The deployment Summary on the right shows the estimated hourly/monthly rate in real time. Click Next.

Step 2: Deployment settings​

  1. Enter a Pod name (lowercase letters, digits, and - only; it must start with a letter and end with a letter or digit).
  2. Choose an image: click + Select Official Templates and pick an official image template published by the administrator. If the template has preset deployment parameters such as ports and a start command, they are filled in automatically.
  3. Click Next.

Step 3: Mount storage (optional)​

You can skip this on your first run. If you need to persist data, select an existing volume here or create one on the spot; see Storage for details. Click Next.

Step 4: Confirm and deploy​

The summary page consolidates the resource, image, parameters, and estimated rental cost. After confirming everything, click Deploy.

Back on the Pods page, deployment is complete once the status changes from Pending to Running.

Step 5: Connect​

Click the newly deployed Pod, open the Connection tab, and click the Web Terminal link to work in the container directly from your browser.

caution

Pods are billed by resource usage hours; credits are deducted continuously while a Pod is running. Remember to delete the Pod when you are done.

Next steps​