> ## Documentation Index
> Fetch the complete documentation index at: https://langchain-zh.cn/llms.txt
> Use this file to discover all available pages before exploring further.

# AWS SageMakerEndpoint 集成

> 使用 LangChain JavaScript 与 AWS SageMakerEndpoint LLM 集成。

LangChain.js 支持与 AWS SageMaker 托管的端点集成。查看 [Amazon SageMaker JumpStart](https://aws.amazon.com/sagemaker/jumpstart/) 以获取可用模型列表，以及如何部署您自己的模型。

## 设置

您需要安装官方的 SageMaker SDK 作为对等依赖项：

```bash npm theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
npm install @aws-sdk/client-sagemaker-runtime
```

<Tip>
  有关安装 LangChain 包的通用说明，请参阅[此部分](/oss/javascript/langchain/install)。
</Tip>

```bash npm theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
npm install @langchain/community @langchain/core
```

## 使用方式

```typescript theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
import {
  SageMakerEndpoint,
  SageMakerLLMContentHandler,
} from "@langchain/community/llms/sagemaker_endpoint";

interface ResponseJsonInterface {
  generation: {
    content: string;
  };
}

// 根据您将使用的模型进行自定义
class LLama213BHandler implements SageMakerLLMContentHandler {
  contentType = "application/json";

  accepts = "application/json";

  async transformInput(
    prompt: string,
    modelKwargs: Record<string, unknown>
  ): Promise<Uint8Array> {
    const payload = {
      inputs: [[{ role: "user", content: prompt }]],
      parameters: modelKwargs,
    };

    const stringifiedPayload = JSON.stringify(payload);

    return new TextEncoder().encode(stringifiedPayload);
  }

  async transformOutput(output: Uint8Array): Promise<string> {
    const response_json = JSON.parse(
      new TextDecoder("utf-8").decode(output)
    ) as ResponseJsonInterface[];
    const content = response_json[0]?.generation.content ?? "";
    return content;
  }
}

const contentHandler = new LLama213BHandler();

const model = new SageMakerEndpoint({
  endpointName: "aws-llama-2-13b-chat",
  modelKwargs: {
    temperature: 0.5,
    max_new_tokens: 700,
    top_p: 0.9,
  },
  endpointKwargs: {
    CustomAttributes: "accept_eula=true",
  },
  contentHandler,
  clientOptions: {
    region: "您的 AWS 端点区域",
    credentials: {
      accessKeyId: "您的 AWS 访问密钥 ID",
      secretAccessKey: "您的 AWS 秘密访问密钥",
    },
  },
});

const res = await model.invoke(
  "你好，我的名字是 John Doe，给我讲一个关于羊驼的笑话 "
);

console.log(res);

/*
  [
    {
      content: "你好，John Doe！这里有一个关于羊驼的笑话：
        为什么羊驼成了园丁？
        因为它擅长羊驼景观设计！"
    }
  ]
 */
```

## 相关链接

* [模型指南](/oss/javascript/langchain/models)

***

<div className="source-links">
  <Callout icon="edit">
    [Edit this page on GitHub](https://github.com/langchain-ai/docs/edit/main/src/i18n\zh-CN\oss\javascript\integrations\llms\aws_sagemaker.mdx) or [file an issue](https://github.com/langchain-ai/docs/issues/new/choose).
  </Callout>

  <Callout icon="terminal-2">
    [Connect these docs](/use-these-docs) to Claude, VSCode, and more via MCP for real-time answers.
  </Callout>
</div>