Azure OpenAI Setup Guide

Here is the guide for using Azure OpenAI models in 16x Prompt:

1. Login to Azure portal
2. Click on "Create a resource"

3. In the create resource page, search for "OpenAI" and click on it

4. Follow the instructions to create a new Azure OpenAI resource
5. Go to Azure home page and click on the "OpenAI" resource you just created, or click on "Azure OpenAI" in Azure services

6. Inside the OpenAI resource, click on "Go to Azure OpenAI Foundry Portal", or the big button at bottom of the page

7. Inside the Azure AI Foundry Portal, you can find your API keys and Azure OpenAl Service endpoint on the home page. Note them down for later use.

8. Scroll down on the left nav bar and click on "Deployments"

9. You can create a new deployment by clicking on "Deploy model" button and follow the instructions to deploy a new model.

10. Keep the deployment name as default (same as the model name)

11. Open 16x Prompt and click on the API model settings button in the bottom right corner

12. Go to "Azure OpenAI API" tab

13. Enter your Azure OpenAI Endpoint name, for example https://xyz.openai.azure.com/
14. Paste the API key into the "API Key" field
15. Enter deployment name, for example gpt-4o into the "Deployment Name" field.
16. Optionally, you can send a second deployment name for comparison, for example gpt-35-turbo.
17. Keep "API Version" as 2024-06-01 unless you have a specific reason to change it.
18. Click on "Close"

Here is an example of using 16x Prompt to compare the results between two deployments (gpt-4o vs gpt-35-turbo):