LangChain is a powerful framework for developing applications powered by language models. Build sophisticated AI apps with ease.
Basic LangChain Setup
from langchain.llms import OpenAI
from langchain.prompts import PromptTemplate
# Initialize LLM
llm = OpenAI(temperature=0.7)
# Create prompt template
template = "Write a {adjective} story about {topic}"
prompt = PromptTemplate(
input_variables=["adjective", "topic"],
template=template
)
# Generate
result = llm(prompt.format(adjective="funny", topic="robots"))
print(result)
```Chains for Multi-Step Tasks
from langchain.chains import LLMChain, SimpleSequentialChain
# First chain: Generate topic
topic_chain = LLMChain(
llm=llm,
prompt=PromptTemplate(
input_variables=["interest"],
template="Suggest a blog topic about {interest}"
)
)
# Second chain: Write outline
outline_chain = LLMChain(
llm=llm,
prompt=PromptTemplate(
input_variables=["topic"],
template="Create an outline for: {topic}"
)
)
# Combine chains
overall_chain = SimpleSequentialChain(
chains=[topic_chain, outline_chain]
)
result = overall_chain.run("artificial intelligence")
```Memory for Conversations
from langchain.memory import ConversationBufferMemory
from langchain.chains import ConversationChain
# Add memory
memory = ConversationBufferMemory()
conversation = ConversationChain(
llm=llm,
memory=memory,
verbose=True
)
# Chat with memory
conversation.predict(input="Hi, my name is Alice")
conversation.predict(input="What's my name?") # Remembers "Alice"
```Document Loaders and Splitting
from langchain.document_loaders import TextLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
# Load document
loader = TextLoader('document.txt')
documents = loader.load()
# Split into chunks
text_splitter = RecursiveCharacterTextSplitter(
chunk_size=1000,
chunk_overlap=200
)
chunks = text_splitter.split_documents(documents)
```Vector Stores for Semantic Search
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores import FAISS
# Create embeddings
embeddings = OpenAIEmbeddings()
# Build vector store
vectorstore = FAISS.from_documents(chunks, embeddings)
# Similarity search
query = "What is machine learning?"
docs = vectorstore.similarity_search(query, k=3)
```Agents for Tool Use
from langchain.agents import load_tools, initialize_agent
from langchain.agents import AgentType
# Load tools
tools = load_tools(["serpapi", "llm-math"], llm=llm)
# Create agent
agent = initialize_agent(
tools,
llm,
agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION,
verbose=True
)
# Use agent
agent.run("What is the population of Tokyo multiplied by 2?")
```LangChain simplifies building complex LLM applications. Start creating today!