Stopping an instance shuts it down and disables connectivity to the main network interface. The operating system and all running processes are stopped, but the instance normal billing cycle continues unless destroyed.

Follow this guide to stop a Vultr Cloud GPU instance using the Vultr Customer Portal, API, or CLI.

Vultr Customer Portal

1. Navigate to Products and click Compute.
2. Click your target instance to open its management page.
3. Click Stop Server on the top-right navigation menu to stop the instance.
4. Click Stop Server in the confirmation prompt to stop the instance.

Vultr API

1. Send a GET request to the List Instances endpoint and note the target instance’s ID in the output.
   console

   $ curl "https://api.vultr.com/v2/instances" \
     -X GET \
     -H "Authorization: Bearer ${VULTR_API_KEY}"
   

2. Send a POST request to the Halt Instances endpoint to stop the instance.
   console

   $ curl "https://api.vultr.com/v2/instances/halt" \
      -X POST \
      -H "Authorization: Bearer ${VULTR_API_KEY}" \
      -H "Content-Type: application/json" \
      --data '{
        "instance_ids" : [
          "your-instance-id"
        ]
      }'

Vultr CLI

1. List all instances in your Vultr account and note the target instance’s ID.
   console

   $ vultr-cli instance list
   

2. Stop the instance.
   console

   $ vultr-cli instance stop <your-instance-id>