Skip to main content
Version: Cloud

Creating an external deployment with Azure OpenAI

Create an Azure OpenAI Deployment using the External Deployment type. In this article, you will find out how to integrate the Azure OpenAI gpt-4o Deployment in Deeploy. All code in this example is also available in this example

Prerequisites

Create a deployment using Azure OpenAI

Use the gpt-4o model that is developed by OpenAI and available on Azure. In order to create an external Deployment in Deeploy you need the following:

  • The model endpoint from a deployed model
  • The API key to communicate with the model

The model inference endpoint and API key

To get a gpt-4o inference endpoint, follow these steps:

  • In Azure AI Foundry create a OpenAI project
  • From the Azure OpenAI model catalog select the gpt-4o model and click deploy
  • Navigate to deployments and select the model that you just deployed
  • You can find the target URI and Key in the Endpoint section

Connecting Azure OpenAI gpt-4o in Deeploy

Connect the Azure OpenAI to Deeploy via the UI or locally with the Python client

Complete the following steps in the Deeploy UI

  1. Login to the Deeploy UI
  2. Navigate to the workspace where you want to add gpt-4o as an external deployment
  3. Click Create > External
  4. Follow the steps, in step 3 you are expected to add:
    • The retrieved inference endpoint (e.g., https://deeploy-openai.openai.azure.com/openai/deployments/gpt-4o/chat/completions?api-version=2024-08-01-preview)
    • Select Custom header as authentication method and use api-key as Custom header
    • Paste the Azure OpenAI Key in the password field
  5. Once the connection check is successful, the deployment can be created.