In this short post I look at Azure AI Studio from Microsoft.
Firstly, I found it a mission to get it working, it is still quite clunky, and doesn't work for all regions. There is a lot of cross-over between Google Vertex AI and Azure AI. So I am not sure who is copying who, but both offer similar interfaces and experiences.
The Nice part about Azure AI is the Microsoft integrations to your domain if you are using Active Directory, which simplifies the security model considerably. But is not a make or break in my mind.
Here is a quick overview of How to install, for a full video check out: https://youtu.be/0JzBzo4KK3o
Step by Step: to deploy a chat web app with an Azure OpenAI Service model.
Prerequisites
An Azure subscription - Create one for free.
Access granted to Azure OpenAI in the desired Azure subscription. Currently, access to this service is granted only by application. You can apply for access to Azure OpenAI by completing the form at https://aka.ms/oai/access. Open an issue on this repo to contact us if you have an issue.
An Azure AI hub resource and project in Azure AI Studio.
Deploy and test a chat model without your data
Follow these steps to deploy a chat model and test it without your data.
Sign in to Azure AI Studio.
Go to your project or create a new project in Azure AI Studio.
Select Build from the top menu and then select Deployments > Create > Real-time endpoint.
4. On the Select a model page, select the model you want to deploy from the Model dropdown. For example, select gpt-35-turbo. Then select Confirm.
5. On the Deploy model page, enter a name for your deployment, and then select Deploy. After the deployment is created, you see the deployment details page. Details include the date you created the deployment and the created date and version of the model you deployed.
6. On the deployment details page from the previous step, select Open in playground.
7. In the playground, make sure that Chat is selected from the Mode dropdown. Select your deployed GPT chat model from the Deployment dropdown.
8. In the System message text box on the Assistant setup pane, provide this prompt to guide the assistant: "You're an AI assistant that helps people find information." You can tailor the prompt for your scenario.
9. Select Apply changes to save your changes, and when prompted to see if you want to update the system message, select Continue.
Deploy your web app
Once you're satisfied with the experience in Azure AI Studio, you can deploy the model as a standalone web application.
Find your resource group in the Azure portal
In this tutorial, your web app is deployed to the same resource group as your Azure AI hub resource. Later you configure authentication for the web app in the Azure portal.
Follow these steps to navigate from Azure AI Studio to your resource group in the Azure portal:
In Azure AI Studio, select Manage from the top menu and then select Details. If you have multiple Azure AI hub resources, select the one you want to use in order to see its details.
In the Resource configuration pane, select the resource group name to open the resource group in the Azure portal.
You should now be in the Azure portal, viewing the contents of the resource group where you deployed the Azure AI hub resource. Keep this page open in a browser tab - you return to it later.
Deploy the web app
Publishing creates an Azure App Service in your subscription. It might incur costs depending on the pricing plan you select. When you're done with your app, you can delete it from the Azure portal.
To deploy the web app:
Complete the steps in the previous section to add your data to the playground. Note You can deploy a web app with or without your own data, but at least you need a deployed model as described in deploy and test a chat model without your data.
Select Deploy > A new web app.
Configure web app authentication
Select the App Service resource from the list of resources in the resource group.
Return to the browser tab containing the Azure portal (or re-open the Azure portal in a new browser tab) and view the contents of the resource group where you deployed the Azure AI hub resource and web app (you might need to refresh the view the see the web app).
From the collapsible left menu under Settings, select Authentication.
Add an identity provider with the following settings:
Identity provider: Select Microsoft as the identity provider. The default settings on this page restrict the app to your tenant only, so you don't need to change anything else here.
Tenant type: Workforce
App registration: Create a new app registration
Name: The name of your web app service
Supported account types: Current tenant - Single tenant
Restrict access: Requires authentication
Unauthenticated requests: HTTP 302 Found redirect - recommended for websites
Kommentare