メインコンテンツまでスキップ

Create an AI Notebook

An AI Notebook is a JupyterLab environment running on dedicated GPU hardware. You select a region and a GPU configuration, and the portal provisions the notebook for you.

This page walks through creating a notebook from start to first access.

1. Open AI Notebook

  1. In the Workspace sidebar, click AI Notebook under Services.
  2. The notebook list shows existing notebooks for your Workspace, one per region.

The AI Notebook list page

2. Create a new notebook

  1. Click Create Notebook.
  2. The create form opens.

The Create Notebook form

3. Select a region

Choose the region where you want the notebook to run:

RegionAvailable GPU
Hà Nội (Southeast Asia)H100 SXM5
Tokyo (Japan)H200 SXM5

The region you select determines which GPU options are available.

注記

Only regions enabled for your Workspace appear here. If a region is missing, contact your Organization Owner.

4. Select a GPU instance

  1. In the GPU Instance section, select a flavor card.
  2. Each card shows the GPU count, VRAM, CPU, RAM, and hourly cost.
  3. Cards marked Out of stock are not available — select a different flavor.

For first-time users, start with the smallest available flavor. You can re-create the notebook with a larger flavor later if needed.

The Summary and Pricing panels on the right update as you make selections.

警告

Charges start when the notebook reaches Running. The pricing panel shows the estimated hourly cost before you confirm.

5. Configure persistent storage

  1. In the Persistent Storage section, enter a disk size in GB. Minimum is 50 GB, maximum is 10,000 GB.
  2. The mount path defaults to /home/user. Change it only if you have a specific requirement.

Persistent storage retains your files when the notebook is stopped or re-created. GPU instance data does not persist between sessions.

6. Create the notebook

  1. Review the Summary and Pricing panels.
  2. Click Create Notebook.

The notebook card appears on the notebook list with status Creating….

A notebook card showing Creating… status

7. Wait for Running status

Creating… means the GPU instance is being provisioned. This typically takes 3–5 minutes.

When provisioning is complete, the status changes to Running. Do not close the page — the status updates automatically.

注記

Creating… is normal. If the status does not change to Running after 10 minutes, see Troubleshooting.

8. Access the notebook

  1. When the status is Running, the Access button becomes available.
  2. Click Access to open JupyterLab in a new tab.

A notebook card in Running status with the Access button


What's next