Azure OpenAI

Azure OpenAI 通过专用部署提供企业级的 OpenAI 模型访问，保障安全性、合规性及服务级别协议（SLA）。

🌐 Azure OpenAI provides enterprise-grade access to OpenAI models through dedicated deployments with security, compliance, and SLA guarantees.

与其他使用固定模型名称的提供商不同，Azure 使用你在 Azure 门户中配置的 部署名称。

🌐 Unlike other providers that have fixed model names, Azure uses deployment names that you configure in the Azure Portal.

Azure 部署的工作原理

🌐 How Azure Deployments Work

Azure 模型 ID 遵循此模式：azure-openai/your-deployment-name

🌐 Azure model IDs follow this pattern: azure-openai/your-deployment-name

部署名称针对你的 Azure 账户是唯一的，并在你在 Azure 门户中创建部署时选择。常见示例：

🌐 The deployment name is specific to your Azure account and chosen when you create a deployment in Azure Portal. Common examples:

• azure-openai/my-gpt4-deployment
• azure-openai/production-gpt-35-turbo
• azure-openai/staging-gpt-4o

设置

🌐 Setup

在 Azure OpenAI Studio 中创建部署。资源名称和 API 密钥可在 Azure 门户的“密钥和终结点”中找到。

🌐 Create deployments in Azure OpenAI Studio. The resource name and API key are in Azure Portal under "Keys and Endpoint".

配置

🌐 Configuration

实例化网关并将其传递给Mastra。提供三种配置模式。

🌐 Instantiate the gateway and pass it to Mastra. Three configuration modes are available.

静态部署

🌐 Static Deployments

从 Azure 门户提供部署名称。

🌐 Provide deployment names from Azure Portal.

import { Mastra } from "@mastra/core";
import { AzureOpenAIGateway } from "@mastra/core/llm";

export const mastra = new Mastra({
  gateways: [
    new AzureOpenAIGateway({
      resourceName: "my-openai-resource",
      apiKey: process.env.AZURE_API_KEY!,
      deployments: ["gpt-4-prod", "gpt-35-turbo-dev"],
    }),
  ],
});

动态发现

🌐 Dynamic Discovery

提供管理 API 凭据。网关查询 Azure 管理 API 以列出部署。

🌐 Provide Management API credentials. The gateway queries Azure Management API to list deployments.

import { Mastra } from "@mastra/core";
import { AzureOpenAIGateway } from "@mastra/core/llm";

export const mastra = new Mastra({
  gateways: [
    new AzureOpenAIGateway({
      resourceName: "my-openai-resource",
      apiKey: process.env.AZURE_API_KEY!,
      management: {
        tenantId: process.env.AZURE_TENANT_ID!,
        clientId: process.env.AZURE_CLIENT_ID!,
        clientSecret: process.env.AZURE_CLIENT_SECRET!,
        subscriptionId: process.env.AZURE_SUBSCRIPTION_ID!,
        resourceGroup: "my-resource-group",
      },
    }),
  ],
});

服务主体需要“Cognitive Services 用户”角色。请参阅 Azure 文档。

🌐 The Service Principal requires "Cognitive Services User" role. See Azure documentation.

手动部署名称

🌐 Manual Deployment Names

仅提供资源名称和 API 密钥。创建代理时请指定部署名称。不要使用 IDE 自动补全。

🌐 Provide resource name and API key only. Specify deployment names when creating agents. No IDE autocomplete.

import { Mastra } from "@mastra/core";
import { AzureOpenAIGateway } from "@mastra/core/llm";

export const mastra = new Mastra({
  gateways: [
    new AzureOpenAIGateway({
      resourceName: "my-openai-resource",
      apiKey: process.env.AZURE_API_KEY!,
    }),
  ],
});

配置参考

🌐 Configuration Reference

选项	类型	是否必填	描述
resourceName	string	是	Azure OpenAI 资源名称
apiKey	string	是	来自“密钥和终结点”的 API 密钥
apiVersion	string	否	API 版本（默认: 2024-04-01-preview）
deployments	string[]	否	静态模式的部署名称
management	object	否	管理 API 凭据
management.tenantId	string	是*	Azure AD 租户 ID
management.clientId	string	是*	服务主体客户端 ID
management.clientSecret	string	是*	服务主体密钥
management.subscriptionId	string	是*	Azure 订阅 ID
management.resourceGroup	string	是*	资源组名称

• 如果提供了 management，则为必填

用法

🌐 Usage

import { Agent } from "@mastra/core/agent";

const agent = new Agent({
  id: "my-agent",
  name: "My Agent",
  instructions: "You are a helpful assistant",
  model: "azure-openai/my-gpt4-deployment"  // Use your Azure deployment name (autocompleted in dev mode)
});

// Generate a response
const response = await agent.generate("Hello!");

// Stream a response
const stream = await agent.stream("Tell me a story");
for await (const chunk of stream) {
  console.log(chunk);
}

请查看 Azure OpenAI 模型可用性 以获取特定地区的选项。

🌐 Check Azure OpenAI model availability for region-specific options.