LangChain architecture overview - Mini Project: Build & Apply

LangChain Architecture Overview

📖 Scenario: You are building a simple LangChain setup to understand its architecture. LangChain helps connect language models with data and tools in a structured way.

🎯 Goal: Create a basic LangChain architecture example with these parts: a language model, a prompt template, a chain that connects them, and a final call to run the chain.

📋 What You'll Learn

Create a variable called llm that holds an OpenAI language model instance with model name 'text-davinci-003'.

Create a variable called prompt that holds a PromptTemplate with input variable 'name' and template 'Hello, {name}!'

Create a variable called chain that holds an LLMChain using the llm and prompt variables.

Call the chain.run method with the argument name='Alice'.

💡 Why This Matters

🌍 Real World

LangChain is used to build applications that combine language models with data sources and tools, like chatbots, question answering, and automation.

💼 Career

Understanding LangChain architecture helps developers create advanced AI-powered applications that integrate language models effectively.

1

Set up the language model

Import OpenAI from langchain.llms and create a variable called llm that holds an OpenAI instance with model_name='text-davinci-003'.

LangChain

# Import OpenAI and create llm with model_name='text-davinci-003'
# Your code here

Need a hint?

Use OpenAI(model_name='text-davinci-003') to create the language model.

2

Create the prompt template

Import PromptTemplate from langchain.prompts and create a variable called prompt that holds a PromptTemplate with input_variables=['name'] and template='Hello, {name}!'.

LangChain

from langchain.llms import OpenAI

llm = OpenAI(model_name='text-davinci-003')
# Import PromptTemplate and create prompt with input_variables=['name'] and template='Hello, {name}!'
# Your code here

Need a hint?

Use PromptTemplate(input_variables=['name'], template='Hello, {name}!') to create the prompt.

3

Create the chain

Import LLMChain from langchain.chains and create a variable called chain that holds an LLMChain using the llm and prompt variables.

LangChain

from langchain.llms import OpenAI
from langchain.prompts import PromptTemplate

llm = OpenAI(model_name='text-davinci-003')
prompt = PromptTemplate(input_variables=['name'], template='Hello, {name}!')
# Import LLMChain and create chain using llm and prompt
# Your code here

Need a hint?

Use LLMChain(llm=llm, prompt=prompt) to create the chain.

4

Run the chain

Call the chain.run method with the argument name='Alice' to get the output.

LangChain

from langchain.llms import OpenAI
from langchain.prompts import PromptTemplate
from langchain.chains import LLMChain

llm = OpenAI(model_name='text-davinci-003')
prompt = PromptTemplate(input_variables=['name'], template='Hello, {name}!')
chain = LLMChain(llm=llm, prompt=prompt)
# Run the chain with name='Alice'
# Your code here

Need a hint?

Use chain.run(name='Alice') to run the chain.