Providers

Azure OpenAI

Enterprise OpenAI deployment on Azure

☁️

Azure OpenAI

Enterprise OpenAI on Azure

Azure OpenAI

Run OpenAI models through Azure for enterprise features: SLAs, compliance, and regional deployment.

Azure OpenAI provides the same models as OpenAI but with enterprise features and Azure integration.


Setup

1. Create Azure OpenAI Resource

  1. Go to Azure Portal
  2. Create an "Azure OpenAI" resource
  3. Deploy a model (e.g., gpt-4o)
  4. Note your endpoint and key

2. Add Environment Variables

# .env.local
AZURE_OPENAI_API_KEY=...
AZURE_OPENAI_ENDPOINT=https://your-resource.openai.azure.com

3. Configure Provider

<YourGPTProvider
  runtimeUrl="/api/chat"
  llm={{
    provider: 'azure',
    model: 'your-deployment-name',  // Your deployment name, not model name
    resourceName: 'your-resource',
  }}
>
  <CopilotChat />
</YourGPTProvider>

Configuration

llm={{
  provider: 'azure',
  model: 'gpt-4o-deployment',      // Your deployment name
  resourceName: 'my-openai',        // Azure resource name
  apiVersion: '2024-02-15-preview', // API version
  temperature: 0.7,
  maxTokens: 4096,
}}

Available Models

Deploy any OpenAI model to your Azure resource:

ModelDeployment NameNotes
GPT-4ogpt-4oLatest multimodal
GPT-4 Turbogpt-4-turbo128K context
GPT-3.5 Turbogpt-35-turboFast, affordable

The model parameter should be your deployment name, not the model name.


Enterprise Features

Regional Deployment

Deploy in your preferred Azure region:

  • US East, West
  • Europe (UK, France, Germany)
  • Asia (Japan, Australia)

Compliance

  • SOC 2, ISO 27001 compliant
  • HIPAA eligible
  • GDPR ready

SLA

  • 99.9% uptime SLA
  • Enterprise support

Runtime Configuration

// app/api/chat/route.ts
import { createRuntime } from '@yourgpt/copilot-sdk-runtime';

const runtime = createRuntime({
  providers: {
    azure: {
      apiKey: process.env.AZURE_OPENAI_API_KEY,
      endpoint: process.env.AZURE_OPENAI_ENDPOINT,
      apiVersion: '2024-02-15-preview',
    },
  },
});

Private Endpoints

For maximum security, use Azure Private Link:

// Traffic stays within Azure network
llm={{
  provider: 'azure',
  model: 'gpt-4o-deployment',
  resourceName: 'my-openai',
  // Uses private endpoint configured in Azure
}}

Pricing

Azure OpenAI uses pay-as-you-go pricing similar to OpenAI, plus Azure infrastructure costs.

Check Azure OpenAI pricing for current rates.


Next Steps

On this page