Azure OpenAI Connection
This topic describes how to authenticate to Azure OpenAI and configure any necessary connection properties in the Azure OpenAI connection connector.
Prerequisites
Prerequisites for using Azure OpenAI services typically include an Azure subscription and proper configuration and authentication settings in Azure to access the API.
To successfully make a call to Azure OpenAI, you need an endpoint and a key. For information on obtaining these details, refer to the topic Retrieve key and endpoint.
To learn more about the prerequisites for using Azure OpenAI and the steps to create a resource and deploy a model, follow the topic Create and Deploy an Azure OpenAI Service resource.
To get the Subscription ID, follow the topic Find your Azure subscription.
To get the Resource Group Name, follow the topic Create a resource.
To get the Directory (Tenant) ID, follow the topic Find your Microsoft Entra tenant.
Connection Configuration
Each connection property available in the Azure OpenAI connection configuration is explained below.
OPEN API DETAILS
To obtain the Service API Key and Azure OpenAI Endpoint, please follow the Prerequisites.
Connection Name
The name of the connection to be created should be provided. This is the name that will display on the list of available connections.
Service API Key
The secret key for accessing the endpoint.
Azure OpenAI End Point
Provide the endpoint of the Azure cognitive service.
MANAGEMENT API DETAILS
If Management API details are provided in your connection configuration, any existing Deployment Name and Base Model associated with your Azure account will automatically be listed in the drop-downs. Otherwise, you will need to provide these details manually.
To obtain these details, please follow the Prerequisites.
Fetch Deployment Details
Gets the deployments associated with the cognitive services accounts.
Subscription ID
The Subscription ID for the cognitive services resource. Tenants can have one or more subscriptions, which are agreements with Microsoft to use cloud services, including Azure. Every Azure resource is associated with a subscription.
Resource Group Name
The Resource Group Name for the Azure cognitive service. A resource group is a container that holds related resources for an Azure solution. The resource group can include all the resources for the solution, or only those resources that you want to manage as a group.
Directory (tenant) ID
The Directory (tenant) ID is a unique identifier for your Azure Active Directory instance. It can be in the form of a domain name or a GUID. Specify the Directory ID associated with your organization’s Azure AD to ensure proper authentication and access to Azure resources.
Application (client) ID
The Application (client) ID is a unique identifier for your Azure application registered in Azure Active Directory. This ID is used to authenticate the application and grant it access to Azure resources. Make sure to use the correct Application ID associated with your application configuration.
Secret Key
The Secret Key generated for your application during the application registration process in Azure AD.
Azure OpenAI Service Account Name
Account name for the Azure cognitive service should be provided.
FETCH DEPLOYMENT DETAILS
After providing all the management API details, click on the “FETCH DEPLOYMENT DETAILS” option to automatically retrieve the OpenAI deployment details, including the deployment name and base model.
OPENAI DEPLOYMENT DETAILS
Deployment Name and Base Model are needed in Azure OpenAI to uniquely identify and customize the behavior of AI models for specific tasks and versions. To obtain these details, please follow the Prerequisites.
If MANAGEMANT API DETAILS are provided, existing Deployment Names and Base Models associated with your Azure account will be listed in the drop-downs; otherwise, you must provide these details manually.
Deployment Name
Enter a unique name for your AI model deployment. You can find existing deployments associated with your Azure account if MANAGEMENT API DETAILS are provided. Otherwise, you can give it manually.
Base Model
Select the base model that serves as the foundation for your AI model’s behavior. Choose a model tailored to your specific use case or task for optimal performance and customization. The base model determines how the AI processes and responds to input data.
You can find existing base models associated with your Azure account if MANAGEMENT API DETAILS are provided. Otherwise, you can give it manually.
Add Configuration: Additional properties can be added using this option as key-value pairs.
After entering all the details, click on the TEST button.
If the connection service identification and authentication details are provided correctly, a success message stating “connection available” is generated.
Click on the CREATE button to save the changes.
If the details are incorrect or the server is down, you will get a message “Connection unavailable”.
If you have any feedback on Gathr documentation, please email us!