Skip to main content
The azure-openai provider supports the Azure OpenAI /chat/completions endpoint.
If you’re using the new Microsoft Foundry portal, use the microsoft-foundry provider instead.

​
Quick Start

client<llm> MyClient {
  provider "azure-openai"
  options {
    resource_name "my-resource-name"
    deployment_id "my-deployment-id"
    api_version "2024-02-01"
    api_key env.AZURE_OPENAI_API_KEY
  }
}
api_version is required. Azure will return “not found” if the version is not specified.

​
Authentication

Set your Azure OpenAI API key as an environment variable: 
export AZURE_OPENAI_API_KEY="your-api-key-here"
Or specify it explicitly in your BAML configuration: 
client<llm> MyClient {
  provider "azure-openai"
  options {
    resource_name "my-resource-name"
    deployment_id "my-deployment-id"
    api_version "2024-02-01"
    api_key env.MY_AZURE_KEY
  }
}

​
Configuration Options

​
BAML-Specific Options

These options modify the API request sent to Azure OpenAI.
​
api_key
string
default:"env.AZURE_OPENAI_API_KEY"
Injected via the API-KEY header.
​
resource_name
string
required
Your Azure OpenAI resource name. Used to construct the base URL.The base URL is constructed as:
https://${resource_name}.openai.azure.com/openai/deployments/${deployment_id}
​
deployment_id
string
required
Your Azure OpenAI deployment ID. This is the name you gave your model deployment in the Azure portal.
​
api_version
string
required
The Azure OpenAI API version. Passed as a query parameter api-version.Common versions:
  • 2024-02-01
  • 2023-12-01-preview
See Azure API versioning for available versions.
​
base_url
string
Alternative to resource_name and deployment_id. Specify the full base URL directly:
client<llm> MyClient {
  provider "azure-openai"
  options {
    base_url "https://my-resource-name.openai.azure.com/openai/deployments/my-deployment-id"
    api_version "2024-02-01"
    api_key env.AZURE_OPENAI_API_KEY
  }
}
​
headers
object
Additional headers to send with requests.
client<llm> MyClient {
  provider "azure-openai"
  options {
    resource_name "my-resource-name"
    deployment_id "my-deployment-id"
    api_version "2024-02-01"
    api_key env.AZURE_OPENAI_API_KEY
    headers {
      "X-My-Header" "my-value"
    }
  }
}

​
Model Parameters

These parameters are passed directly to the Azure OpenAI API.
For reasoning models like o1 or o1-mini, use max_completion_tokens instead of max_tokens. Set max_tokens to null.See the OpenAI API documentation for details.
client<llm> AzureO1 {
  provider "azure-openai"
  options {
    resource_name "my-resource"
    deployment_id "o1-mini"
    api_version "2024-02-01"
    max_tokens null
    max_completion_tokens 4096
  }
}
Common parameters:
  • temperature - Controls randomness (0-2)
  • max_tokens - Maximum tokens to generate
  • top_p - Nucleus sampling parameter
  • frequency_penalty - Reduces repetition (-2.0 to 2.0)
  • presence_penalty - Encourages new topics (-2.0 to 2.0)
For all options, see the Azure OpenAI API documentation.

​
Finding Your Configuration

In the Azure Portal:
  1. Go to your Azure OpenAI resource
  2. Navigate to “Keys and Endpoint”
  3. Find:
    • Resource name: In the endpoint URL (e.g., https://YOUR-RESOURCE.openai.azure.com/)
    • API key: Under “Key 1” or “Key 2”
  4. Navigate to “Model deployments” to find your deployment_id

​
Features

  • Streaming: Supported for real-time response generation
  • Multimodal: Support depends on your deployed model
  • Azure Integration: Works with Azure AD authentication and private endpoints
  • Regional Deployment: Deploy in your preferred Azure region

​
Do Not Set

​
messages
DO NOT USE
BAML automatically constructs this from your prompt.
​
stream
DO NOT USE
BAML automatically sets this based on how you call the client in your code.

Build docs developers (and LLMs) love

Get started for freeTalk to us