---
url: /v6.2/en-us/azure-openai.md
---
# Connect to Microsoft Azure OpenAI Service

1. Log in to https://ai.azure.com
2. Click `Projects` ![](/images/copilot/azure-openai-project.png)
3. Click `Deployments` ![](/images/copilot/azure-openai-deploy.png)
4. Click `Models` ![](/images/copilot/azure-openai-model.png)
5. Copy the `Endpoint` ![](/images/copilot/azure-openai-endpoint.png)
6. In the HENGSHI Intelligent Query Assistant's model configuration:
   * set the `Model Address` to the `Endpoint`, for example, `https://openai-copilot.openai.azure.com/openai/deployments/gpt-4-32k-0314/versions/2023-03-15-preview`
   * set the `Key` to your `API Key`, for example, `sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx`
   * set the `Model` to the corresponding `gpt-4-32k-0314` in the endpoint address.
7. Click `Test Model Connection`. If it displays `Connection Successful`, the configuration is correct. ![Test Model Connection](/images/copilot/test-connection.png)
8. Finally, click `Save`.