一、LangChain简介

LangChain是一个用于构建LLM应用的开发框架，GitHub 118K+ Stars。提供Chain、Agent、Memory、RAG等模块化组件。

核心模块

二、安装部署

基本安装

基本安装

pip install langchain
OpenAI集成

pip install langchain-openai
所有集成

pip install langchain[all]

可选集成

向量数据库

pip install langchain-chroma
pip install langchain-pinecone
Agent工具

pip install langchain-community
其他

pip install langchain-anthropic
pip install langchain-deepseek

三、使用教程

基本对话

from langchain_openai import ChatOpenAI
llm = ChatOpenAI(
    model="gpt-4",
    api_key="sk-xxx"
)
response = llm.invoke("Hello, how are you?")
print(response.content)

Chain示例

from langchain_openai import ChatOpenAI
from langchain.prompts import PromptTemplate
from langchain.chains import LLMChain
llm = ChatOpenAI(model="gpt-4")
prompt = PromptTemplate(
    input_variables=["topic"],
    template="Write a short story about {topic}"
)
chain = LLMChain(llm=llm, prompt=prompt)
result = chain.run("a robot learning to paint")
print(result)

RAG示例

from langchain_openai import OpenAIEmbeddings, ChatOpenAI
from langchain_chroma import Chroma
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.chains import RetrievalQA
加载文档

from langchain_community.document_loaders import TextLoader
loader = TextLoader("document.txt")
documents = loader.load()
分割

splitter = RecursiveCharacterTextSplitter(chunk_size=1000)
docs = splitter.split_documents(documents)
向量化

embeddings = OpenAIEmbeddings()
vectorstore = Chroma.from_documents(docs, embeddings)
创建RAG链

llm = ChatOpenAI(model="gpt-4")
qa = RetrievalQA.from_chain_type(llm=llm, retriever=vectorstore.as_retriever())
问答

result = qa.invoke("What is the document about?")

四、支付相关

LangChain免费，但需要配置LLM API：

API配置

OpenAI：

import os
os.environ["OPENAI_API_KEY"] = "sk-xxx"

DeepSeek（推荐低成本）：

import os
os.environ["OPENAI_API_KEY"] = "your-deepseek-key"
os.environ["OPENAI_API_BASE"] = "https://api.deepseek.com/v1"
from langchain_openai import ChatOpenAI
llm = ChatOpenAI(
    model="deepseek-chat",
    api_key=os.environ["OPENAI_API_KEY"],
    base_url=os.environ["OPENAI_API_BASE"]
)

成本优化