Azure AI Search provides a fast, scalable, reliable vector search service that you can extend with RAG and other AI services.

In this article, I will show how to create an Azure AI Search service.

Navigate to the Azure Portal and log in.

Click the [Create a resource] button (Fig. 1) and search for "ai search" or "azure ai search," as shown in Fig. 2.

Fig. 1

Fig. 2

From the list of results, select the [Create] button in the "Azure AI Search" panel (Fig. 3) to expand the menu, and select the [Create] option, as shown in Fig. 4.

Fig. 3

Fig. 4

The "Create a search service" dialog displays with the "Basics" tab selected, as shown in Fig. 5.

Fig. 5

At the "Subscription" dropdown, select the subscription in which you want to create this AI Search service. Many of you will have only one subscription, so you will not need to choose anything here.

At the "Resource group" field, select the resource group in which you want to create the Search service, or click the "Create new" link to create a new resource group in which to add the Search service. A resource group is a logical grouping of Azure resources you want to manage together.

At the "Service name" field, enter a unique name for this Search service.

At the "Location" dropdown, select an Azure region in which to create the Search service. Consider the location of the people and services using this service to minimize latency.

The "Pricing tier" field defaults to "Standard." If you want to change this, click the "Change pricing tier" link and select an appropriate tier from the list of options, as shown in Fig. 6.

Fig. 6

These pricing tiers are listed in ascending order of price and capacity. You should select one that meets your needs, but resist paying for more than you need.

Fig. 7 shows the "Scale" tab. It is unnecessary to change anything on this tab, but it allows you to add more Replicas and partitions. Increasing Replicas increases the availability of the service, while increasing Partitions increases the capacity of the service. You should set the Replicas to at least 3 for production environments to achieve high availability for read and write operations.

Fig. 7

Fig. 8 shows the "Networking" tab. It is unnecessary to change anything on this tab, but it allows you to restrict access to the account to specific networks and configure private endpoints for the account.

Fig. 8

Fig. 9 shows the "Tags" tab. It is not necessary to change anything on this tab, but you can apply name-value pairs to this resource that you may use to filter or sort your reports.

Fig. 9

Fig. 10 shows the "Review + create" tab. If you made any errors, such as leaving a required field empty or selecting an inconsistent combination of options, these errors will be listed here, and you will need to correct them before you can proceed.

Fig. 10

After correcting any errors, click the [Create] button (Fig. 11) to start creating the Azure AI Search service.

Fig. 11

After a short time, a confirmation message like the one in Fig. 12 will display, indicating that the Search service has been created.

Fig. 12

Click the [Go to resource] button (Fig. 13) to show the "Overview" blade of the newly created Azure AI Search service, as shown in Fig. 14.

Fig. 13

Fig. 14

This article showed you how to create an Azure AI Search service.