PerplexitySearchRetriever 集成

使用 LangChain JavaScript 集成 PerplexitySearchRetriever retriever。

The Perplexity Search API returns real-time, grounded web search results that you can drop directly into a retrieval pipeline.

概述

This will help you get started with PerplexitySearchRetriever retrieval. For detailed documentation of all PerplexitySearchRetriever features and configurations head to the API reference.

集成详情

Retriever	Source	Package	PY support
PerplexitySearchRetriever	The web, via the Perplexity Search API.	@langchain/perplexity	✅

设置

You will need to populate a PERPLEXITY_API_KEY environment variable with your Perplexity API key, or pass it into the constructor as apiKey. Get a key from the Perplexity API key dashboard.

process.env.PERPLEXITY_API_KEY = "your-api-key";

If you want to get automated tracing from individual queries, you can also set your LangSmith API key by uncommenting below:

// process.env.LANGSMITH_API_KEY = "<YOUR API KEY HERE>";
// process.env.LANGSMITH_TRACING = "true";

安装

This retriever lives in the @langchain/perplexity package:

npm

npm install @langchain/perplexity @langchain/core

实例化

import { PerplexitySearchRetriever } from "@langchain/perplexity";

const retriever = new PerplexitySearchRetriever({
  maxResults: 5,
  // searchDomainFilter: ["wikipedia.org"],
  // searchRecencyFilter: "week",
});

Constructor options include maxResults, country, searchDomainFilter, searchRecencyFilter, searchAfterDate, and searchBeforeDate. See the Perplexity Search API parameters for details.

用法

const query = "Who won the most recent FIFA World Cup?";

const docs = await retriever.invoke(query);

Each returned Document has the search result snippet as pageContent and { title, url, date, lastUpdated } in metadata.

Use within a chain

Like other retrievers, PerplexitySearchRetriever can be incorporated into LLM applications via chains.

import { ChatPromptTemplate } from "@langchain/core/prompts";
import { ChatAnthropic } from "@langchain/anthropic";
import {
  RunnablePassthrough,
  RunnableSequence,
} from "@langchain/core/runnables";
import { StringOutputParser } from "@langchain/core/output_parsers";
import { Document } from "@langchain/core/documents";

const prompt = ChatPromptTemplate.fromTemplate(`
Answer the question based only on the context provided.

Context: {context}

Question: {question}`);

const llm = new ChatAnthropic({ model: "claude-3-5-haiku-latest" });

const formatDocs = (docs: Document[]) =>
  docs.map((doc) => doc.pageContent).join("\n\n");

const ragChain = RunnableSequence.from([
  {
    context: retriever.pipe(formatDocs),
    question: new RunnablePassthrough(),
  },
  prompt,
  llm,
  new StringOutputParser(),
]);

await ragChain.invoke("Who won the most recent FIFA World Cup?");

API 参考

有关所有 PerplexitySearchRetriever 功能和配置的详细文档，请前往 API 参考。

---

将这些文档连接到 Claude、VSCode 等工具，通过 MCP 获取实时答案。

在 GitHub 上编辑此页面 或 提交 issue。