Getting Started with LangChain: Build Your First AI Application

LangChain is currently the most popular AI application development framework. This article will guide you from scratch to build a simple RAG Q&A system.

What is LangChain?

LangChain is a framework for building AI applications. It provides:

• Chaining: Connect multiple AI steps in a sequence
• RAG Support: Connect to external data sources
• Agent Capabilities: Let AI make autonomous decisions
• Tool Integration: Call external APIs and tools

Installation

pip install langchain langchain-openai chromadb

Building a RAG Q&A System

1. Prepare Documents

from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_community.vectorstores import Chroma
from langchain_openai import OpenAIEmbeddings

# Prepare your documents
docs = [
    "Ciyuano Ciyuano is a OpenAI compatible API API relay service...",
    "supports DeepSeek、GLM、Qwen and other domestic LLMs...",
    "Usage：Modify base_url and api_key that's it...",
]

# Split documents
splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
chunks = splitter.create_documents(docs)

# Create vector database
vectorstore = Chroma.from_documents(chunks, OpenAIEmbeddings(
    base_url="https://www.ciyuano.com/v1",
    api_key="sk-relay-Your API key"
))

2. Build the Q&A Chain

from langchain.chains import RetrievalQA
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="deepseek-v4-flash",
    base_url="https://www.ciyuano.com/v1",
    api_key="sk-relay-Your API key"
)

qa_chain = RetrievalQA.from_chain_type(
    llm=llm,
    retriever=vectorstore.as_retriever(),
)

# Query
result = qa_chain.invoke({"query": "Which models does Ciyuano support？"})
print(result["result"])

Core Concepts

Prompt Template

Prompt template, used to build structured prompts.

Chain

Connect multiple steps to form a complete processing flow.

Retriever

Retrieve relevant documents from the vector database.

Agent

Let the AI autonomously decide which tools to use to complete tasks.

Next Steps

• Try connecting to different models
• Build more complex Agents
• Integrate external tools (search, database, etc.)
• Deploy as a Web service

Resources

• LangChain Official Documentation
• Ciyuano Ciyuano API Documentation