How to create a AI foundry project under my sponsor subscription

Logan Yang 25 Reputation points
2025-03-17T05:45:48.9033333+00:00

Hi,
I'm trying to use my Microsoft Azure Sponsorship credits for the AI Foundry serverless models. I can't seem to create a project and deployment using this Microsoft Azure Sponsorship.

Plus, the AI foundry pages are quite confusing on its own. I was used to the Azure OpenAI pages last year but now it's very complex. For example, there are "Azure OpenAI" and "AI inference" from a dropdown, and I was not able to find a CURL example for a REST API, there's only code samples.

Please help, thanks!

Logan

Azure AI services
Azure AI services
A group of Azure services, SDKs, and APIs designed to make apps more intelligent, engaging, and discoverable.
3,602 questions
{count} votes

Accepted answer
  1. SriLakshmi C 6,010 Reputation points Microsoft External Staff Moderator
    2025-03-17T09:48:02.55+00:00

    Hello Logan Yang,

    Certain restrictions may apply depending on your sponsorship type or available credits. For instance, some enterprise agreements or promotional credits may not include access to all services. Make sure you have the required permissions for your sponsor subscription.

    To create an AI Foundry Project under your sponsor subscription in Azure, please follow these steps:

    1: Sign in to Azure Portal Go to the Azure Portal and sign in with your credentials. Ensure that you have selected the correct sponsor subscription.

    2: Navigate to AI Foundry (Azure AI Studio) Open Azure AI Studio (ai.azure.com) and click on "Create Project" under AI Foundry.

    3: Configure Your AI Foundry Project Choose a meaningful Project Name, select your sponsor subscription, and either create or use an existing Resource Group. Select a Region closest to your users. Under Services to Include, choose AI services such as Azure OpenAI, Azure ML, and Cognitive Services (Speech, Vision, Language). Additionally, configure the appropriate Storage & Compute resources for your project.

    4: Set Up AI Models Add pre-trained AI models like OpenAI (GPT), Vision, and Speech. Configure API endpoints and authentication and enable fine-tuning if needed.

    5: Review & Deploy Click "Review + Create" to validate your configuration. Once confirmed, click "Create" to deploy your AI Foundry Project.

    6: Access & Manage After deployment, navigate to Azure AI Studio to access your AI Foundry project. You can use SDKs or APIs to integrate AI services into your applications efficiently.

    Yes, the new AI Foundry experience in Azure AI Studio has introduced a more complex interface compared to the previous Azure OpenAI pages. Previously, could directly access the Azure OpenAI Playground, but now, everything is integrated into AI Foundry, making navigation more challenging.

    In the new Azure AI Studio, multiple dropdown options are available: Azure OpenAI for accessing GPT models (ChatGPT, GPT-4, fine-tuning, embeddings, etc.), AI Inference for running foundation and custom models efficiently, and Azure ML for training and deploying machine learning models. If you were accustomed to the older Azure OpenAI Service, this structural change might feel unfamiliar.

    1.Where to Find a cURL Example for REST API Calls?

    Microsoft has changed the way API examples are presented. Instead of explicitly providing cURL commands, they now mostly offer code samples in Python, C#, Node.js, and other languages. However, you can still generate a cURL command manually.

    2.How to Generate a cURL Command for REST API Calls in AI Foundry:

    • Go to Azure AI Studio.
    • Select "Azure OpenAI" from the dropdown in the left panel.
    • Click on "Deployments" → Select your model (e.g., GPT-4).
    • Click "Use in Code" (top-right button).
    • Review the provided Python, JavaScript, or C# samples.
    • Manually construct the cURL command using your API key and endpoint.

    3.If you need a cURL example, you can manually construct it as follows:

    curl -X POST "https://YOUR_AZURE_OPENAI_INSTANCE.openai.azure.com/openai/deployments/YOUR_DEPLOYMENT_NAME/chat/completions?api-version=2024-02-15-preview" \
    -H "Content-Type: application/json" \
    -H "api-key: YOUR_AZURE_OPENAI_API_KEY" \
    -d '{
      "messages": [
        {"role": "system", "content": "You are a helpful AI assistant."},
        {"role": "user", "content": "What is AI Foundry?"}
      ],
      "temperature": 0.7,
      "max_tokens": 100
    }'
    

    Attaching a document for your reference Azure OpenAI Service REST API reference.

    I hope this helps. And, if you have any further query do let us know.

    Thank you!

    1 person found this answer helpful.
    0 comments No comments

1 additional answer

Sort by: Most helpful
  1. Amira Bedhiafi 33,071 Reputation points Volunteer Moderator
    2025-03-17T09:30:01.87+00:00

    Hello Logan,

    Thank you for posting your question on Microsoft Learn.

    To create an AI Foundry project under your Microsoft Azure Sponsorship subscription :

    First, go to the Azure Portal and log in with your credentials.

    In the search bar at the top of the Azure Portal, type "AI Services" and select it from the dropdown.

    Click on the "Create" button to start creating a new AI service resource.

    From the list of available AI services, find and select "AI Foundry".

    Make sure that your Microsoft Azure Sponsorship subscription is selected in the subscription dropdown.

    Then you need to configure the AI Foundry Project !

    • Create a new resource group or select an existing one.
    • Choose the region where you want to deploy your AI Foundry project.
    • Select the pricing tier that fits your needs. Ensure that it aligns with your sponsorship credits.

    Review all the settings and click on "Create" to deploy your AI Foundry project.

    Once the deployment is complete, go to the resource page of your AI Foundry project.

    Look for the "API Documentation" or "Quickstart" section. This should provide you with the necessary endpoints and authentication details.

    If you can't find a CURL example directly, you can often infer it from the code samples provided. For example, if you see a Python example using requests, you can convert it to a CURL command.

    Assuming you have an endpoint and an API key, a typical CURL command might look like this:

    
    curl -X POST "https://<your-endpoint>.azurewebsites.net/api/v1/predict" \
    
    -H "Content-Type: application/json" \
    
    -H "Authorization: Bearer <your-api-key>" \
    
    -d '{"input": "your input data"}'
    
    0 comments No comments

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.