Azure OpenAI LLM Provider

The Azure OpenAI provider connects Beluga AI to OpenAI models hosted on Microsoft Azure. It uses the same OpenAI-compatible API format with Azure-specific authentication (api-key header) and URL structure (per-deployment endpoints with api-version query parameter).

Choose Azure OpenAI when your organization requires enterprise compliance controls, Azure Active Directory integration, virtual network isolation, or data residency guarantees. Azure provides the same OpenAI models with enterprise-grade SLAs and private networking.

Installation

go get github.com/lookatitude/beluga-ai/llm/providers/azure

Configuration

Field	Required	Default	Description
Model	No	"gpt-4o"	Model identifier
APIKey	Yes	—	Azure API key
BaseURL	Yes	—	Azure deployment endpoint (see below)
Timeout	No	30s	Request timeout

Provider-specific options (via Options map):

Key	Default	Description
api_version	"2024-10-21"	Azure API version string

BaseURL format:

https://{resource-name}.openai.azure.com/openai/deployments/{deployment-name}

Environment variables:

Variable	Maps to
AZURE_OPENAI_API_KEY	APIKey

Basic Usage

package main

import (
    "context"
    "fmt"
    "log"
    "os"

    "github.com/lookatitude/beluga-ai/config"
    "github.com/lookatitude/beluga-ai/llm"
    "github.com/lookatitude/beluga-ai/schema"
    _ "github.com/lookatitude/beluga-ai/llm/providers/azure"
)

func main() {
    model, err := llm.New("azure", config.ProviderConfig{
        Model:   "gpt-4o",
        APIKey:  os.Getenv("AZURE_OPENAI_API_KEY"),
        BaseURL: "https://myresource.openai.azure.com/openai/deployments/my-gpt4o",
        Options: map[string]any{
            "api_version": "2024-10-21",
        },
    })
    if err != nil {
        log.Fatal(err)
    }

    msgs := []schema.Message{
        schema.NewSystemMessage("You are a helpful assistant."),
        schema.NewHumanMessage("What is the capital of France?"),
    }

    resp, err := model.Generate(context.Background(), msgs)
    if err != nil {
        log.Fatal(err)
    }

    fmt.Println(resp.Text())
}

Streaming

for chunk, err := range model.Stream(context.Background(), msgs) {
    if err != nil {
        log.Fatal(err)
    }
    fmt.Print(chunk.Delta)
}
fmt.Println()

Advanced Features

Tool Calling

Azure OpenAI supports the same tool calling capabilities as OpenAI:

tools := []schema.ToolDefinition{
    {
        Name:        "get_weather",
        Description: "Get current weather for a location",
        InputSchema: map[string]any{
            "type": "object",
            "properties": map[string]any{
                "location": map[string]any{
                    "type":        "string",
                    "description": "City name",
                },
            },
            "required": []any{"location"},
        },
    },
}

modelWithTools := model.BindTools(tools)
resp, err := modelWithTools.Generate(ctx, msgs, llm.WithToolChoice(llm.ToolChoiceAuto))

Structured Output

resp, err := model.Generate(ctx, msgs,
    llm.WithResponseFormat(llm.ResponseFormat{Type: "json_object"}),
)

Generation Options

resp, err := model.Generate(ctx, msgs,
    llm.WithTemperature(0.7),
    llm.WithMaxTokens(2048),
    llm.WithTopP(0.9),
)

How Azure Authentication Works

Azure OpenAI uses a different authentication mechanism than standard OpenAI. The provider automatically handles this by:

1. Setting the api-key HTTP header (instead of Authorization: Bearer)
2. Appending the api-version query parameter to every request
3. Removing the default Authorization header set by the OpenAI SDK

This is transparent to the caller. You only need to provide the correct APIKey and BaseURL.

Error Handling

resp, err := model.Generate(ctx, msgs)
if err != nil {
    // Errors are wrapped with the "openaicompat:" prefix
    log.Fatal(err)
}

Direct Construction

import "github.com/lookatitude/beluga-ai/llm/providers/azure"

model, err := azure.New(config.ProviderConfig{
    Model:   "gpt-4o",
    APIKey:  os.Getenv("AZURE_OPENAI_API_KEY"),
    BaseURL: "https://myresource.openai.azure.com/openai/deployments/my-gpt4o",
})