exa-rag

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Exa RAG Integration

Exa RAG集成

Quick Reference

快速参考

Topic	When to Use	Reference
LangChain	Building RAG chains with LangChain	langchain.md
LlamaIndex	Using Exa as a LlamaIndex data source	llamaindex.md
Vercel AI SDK	Adding web search to Next.js AI apps	vercel-ai.md
MCP & Tools	Claude MCP server, OpenAI tools, function calling	mcp-tools.md

主题	适用场景	参考文档
LangChain	使用LangChain构建RAG链	langchain.md
LlamaIndex	将Exa用作LlamaIndex数据源	llamaindex.md
Vercel AI SDK	为Next.js AI应用添加网页搜索功能	vercel-ai.md
MCP & 工具	Claude MCP服务器、OpenAI工具、函数调用	mcp-tools.md

Essential Patterns

核心模式

LangChain Retriever

LangChain检索器

python

from langchain_exa import ExaSearchRetriever

retriever = ExaSearchRetriever(
    exa_api_key="your-key",
    k=5,
    highlights=True
)

docs = retriever.invoke("latest AI research papers")

python

from langchain_exa import ExaSearchRetriever

retriever = ExaSearchRetriever(
    exa_api_key="your-key",
    k=5,
    highlights=True
)

docs = retriever.invoke("latest AI research papers")

LlamaIndex Reader

LlamaIndex阅读器

python

from llama_index.readers.web import ExaReader

reader = ExaReader(api_key="your-key")
documents = reader.load_data(
    query="machine learning best practices",
    num_results=10
)

python

from llama_index.readers.web import ExaReader

reader = ExaReader(api_key="your-key")
documents = reader.load_data(
    query="machine learning best practices",
    num_results=10
)

Vercel AI SDK Tool

Vercel AI SDK工具

typescript

import { exa } from "@agentic/exa";
import { createOpenAI } from "@ai-sdk/openai";
import { generateText } from "ai";

const result = await generateText({
  model: openai("gpt-4"),
  tools: { search: exa.searchAndContents },
  prompt: "Search for the latest TypeScript features",
});

typescript

import { exa } from "@agentic/exa";
import { createOpenAI } from "@ai-sdk/openai";
import { generateText } from "ai";

const result = await generateText({
  model: openai("gpt-4"),
  tools: { search: exa.searchAndContents },
  prompt: "Search for the latest TypeScript features",
});

OpenAI-Compatible Endpoint

OpenAI兼容端点

python

from openai import OpenAI

client = OpenAI(
    base_url="https://api.exa.ai/v1",
    api_key="your-exa-key"
)

response = client.chat.completions.create(
    model="exa",
    messages=[{"role": "user", "content": "What are the latest AI trends?"}]
)

python

from openai import OpenAI

client = OpenAI(
    base_url="https://api.exa.ai/v1",
    api_key="your-exa-key"
)

response = client.chat.completions.create(
    model="exa",
    messages=[{"role": "user", "content": "What are the latest AI trends?"}]
)

Integration Selection

集成方案选择

Framework	Best For	Key Feature
LangChain	Complex chains, agents	ExaSearchRetriever, tool integration
LlamaIndex	Document indexing, Q&A	ExaReader, query engines
Vercel AI SDK	Next.js apps, streaming	Tool definitions, edge-ready
OpenAI Compat	Drop-in replacement	Minimal code changes
Claude MCP	Claude Desktop, Claude Code	Native tool calling

框架	最佳适用场景	核心特性
LangChain	复杂链、Agent	ExaSearchRetriever、工具集成
LlamaIndex	文档索引、问答系统	ExaReader、查询引擎
Vercel AI SDK	Next.js应用、流式处理	工具定义、边缘就绪
OpenAI兼容	直接替换方案	代码改动最小
Claude MCP	Claude桌面端、Claude代码工具	原生工具调用

Common Mistakes

常见误区

Not using highlights for RAG - Full text wastes context; use
```
highlights=True
```
for relevant snippets
Missing source attribution - Always include
```
result.url
```
in citations for grounded responses
Ignoring summaries -
```
summary=True
```
provides concise context without full page overhead
Over-fetching results - Start with 3-5 results; more isn't always better for RAG quality
Not filtering domains - Use
```
include_domains
```
to limit to authoritative sources
Skipping date filters - For current events, always add
```
start_published_date
```
to avoid stale info
Forgetting async patterns - Use async retrievers in production for better throughput

未在RAG中使用高亮功能 - 全文内容会浪费上下文；使用
```
highlights=True
```
获取相关片段
缺失来源归因 - 为保证基于事实的响应，务必在引用中包含
```
result.url
```
忽略摘要功能 -
```
summary=True
```
可提供简洁上下文，避免全页内容的开销
过度获取结果 - 初始设置3-5条结果即可；结果数量过多并不一定提升RAG质量
未过滤域名 - 使用
```
include_domains
```
限制为权威来源
跳过日期过滤 - 针对时事内容，务必添加
```
start_published_date
```
以避免过时信息
忘记异步模式 - 生产环境中使用异步检索器可提升吞吐量