Skip to content

Create an API Proxy for an Azure Open AI API

Follow the steps below to create an AI API:

1. Access the API Platform Console

  1. Sign in to the API Platform Console and select your project.

2. Create an Azure Open AI API

  1. Click +Create.
  2. Navigate to Create an API Proxy for Third-Party APIs (Egress) and click AI APIs.

  3. Select the Azure OpenAI Provider.

    Create AI API Proxy page showing AWS Bedrock, Azure OpenAI, Anthropic Claude, Mistral, and OpenAI provider cards

  4. Configure the AI API details, such as:

    • API Name
    • Identifier
    • Version
    • AI Provider
    • Target Endpoint

    Create AI API Proxy form with Azure OpenAI Service API name, identifier, version, base path, description, type, network visibility, and target fields

  5. Click Create to finalize the AI API creation.

Once created, the overview page of the new API appears.

3. Configure Backend Settings

  1. In the left navigation menu, click Develop, then select Policy.

    Policy page showing API Proxy Contract with POST routes and Service Contract with Azure OpenAI endpoint

  2. Click endpoint Endpoint Configurations Policy page with Endpoint Configuration tooltip highlighted on the Azure OpenAI Service Contract endpoint

  3. Configure Endpoint Configurations to handle backend security authentication and authorization.

    API Key Header API Key
    api-key xxxxxxxxxxxxxxxxx

    Endpoint Configuration dialog showing Azure OpenAI endpoint URL with api-key header and API key input fields

4. Deploy to development and Production Environments

  1. In the left navigation menu, click Deploy, Deploy to Development environment

    Deploy page showing Build Area, Development environment with Active deployment status, and Production not yet deployed

  2. Once it is deploy to development you can test it from Test Console.

  3. When you need to promote to production environment you have to provide endponit credentials for production setup.

    Deploy page with Development and Production both Active, and Configurations panel showing api-key header and API key fields

Next Steps

By following these steps, you can efficiently integrate Open AI services into your API Platform environment and manage them with ease.