Azure OpenAI GPT-4o Structured Outputs with .NET

The goal of this sample is to demonstrate how to leverage the Azure OpenAI GPT-4o structured outputs feature by generating valid OpenAI JSON schemas for a given Type in .NET.

How it works

Important

The original vision was to leverage the JsonSchemaExporter in .NET 9 to generate JSON schemas. However, the generated schemas are not directly compatible with the OpenAI SDK. The current implementation is a workaround to generate valid schemas for structured outputs.

The OpenAIJsonSchema class provides a mechanism to generate a valid OpenAI JSON schema for any given model Type. The class uses reflection to traverse the properties of the model Type and generate the schema based on the property types.

The StructuredOutputsExtensions class provides a generic CreateJsonSchemaFormat method that can be passed a model object type and will return the ChatResponseFormat object that can be used by the OpenAI SDK. This method uses the OpenAIJsonSchema.For method to generate the schema for the model object type.

In addition to the generic CreateJsonSchemaFormat method, a generic CompleteChat and CompleteChatAsync extension is provided for the ChatClient that will return a wrapped ChatCompletion object as ParsedChatCompletion that includes a Parsed property containing the deserialized model object.

Example

class Family
{
    public List<Person> Parents { get; set; }
    public List<Person>? Children { get; set; }

    class Person
    {
        public string Name { get; set; }
        public int Age { get; set; }
    }
}

ChatCompletionOptions options = new()
{
    ResponseFormat = StructuredOutputsExtensions.CreateJsonSchemaFormat<Family>("family", jsonSchemaIsStrict: true),
    MaxOutputTokenCount = 4096,
    Temperature = 0.1f,
    TopP = 0.1f
};

List<ChatMessage> messages =
[
    new SystemChatMessage("You are an AI assistant that creates families."),
    new UserChatMessage("Create a family with 2 parents and 2 children.")
];

ParsedChatCompletion<Family?> completion = chatClient.CompleteChat(messages, options);
Family? family = completion.Parsed;

You can also provide descriptions for schema properties by using the DescriptionAttribute:

class Person
{
    [Description("The name of the person.")]
    public string Name { get; set; }

    [Description("The age of the person.")]
    public int Age { get; set; }
}

These are included in the generated schema:

{
  "type": "object",
  "properties": {
    "Name": {
      "type": "string",
      "description": "The name of the person."
    },
    "Age": {
      "type": "integer",
      "description": "The age of the person."
    }
  }
}

Running the sample

• Install the latest .NET 9 SDK.
• Install PowerShell Core.
• Install the Azure CLI.
• Install Visual Studio Code.

Additionally, you will require:

• An Azure subscription. If you don't have an Azure subscription, create an account.

To setup a local development environment, follow these steps:

Note

For the most optimal sample experience, it is recommended to deploy the necessary infrastructure in a region that supports GPT-4o version 2024-08-06. Find out more about region availability for the GPT-4o model.

az login

./Setup-Environment.ps1 -DeploymentName <UniqueDeploymentName> -Location <AzureRegion> -SkipInfrastructure $false

The script will deploy the following resources to your Azure subscription:

• Azure AI Services, a managed service for all Azure AI Services, including Azure OpenAI.
  • Note: GPT-4o will be deployed as Global Standard with 10K TPM quota allocation. This can be adjusted based on your quota availability in the main.bicep file.

Note

Resources are secured by default with Microsoft Entra ID using Azure RBAC. Your user client ID will be added with the necessary least-privilege roles to access the resources created.

After the script completes, a .env file will be crated in the src/StructuredOutputs folder and you can run the sample project by following these steps:

cd src/StructuredOutputs
dotnet run

Resources

• Azure OpenAI GPT-4o Structured Outputs Supported Schemas and Limitations
• How to use chat completions with structured outputs in OpenAI .NET SDK
• .NET 9 JSON Schema Exporter

License

This project is licensed under the MIT License - see the LICENSE file for details.