How to Install and Configure LangChain for LLM Development

Chapter 1: Installation and Environment Setup

Welcome to the first chapter of the LangChain tutorial. Before building powerful LLM applications, you must install LangChain and configure the development environment correctly.

LangChain Installation

Since version v0.1.0, LangChain uses a modular package structure (e.g., langchain-core, langchain-community), but the tutorial recommends installing the umbrella package langchain to simplify usage.

1. Install LangChain

pip install langchain

This single command provides core abstractions, orchestration logic, Chains, Agents, and Retrieval‑Augmented Generation (RAG) features.

2. Install Model‑Specific Integration Packages

Integrations for particular models or components are released as separate packages. For OpenAI models: pip install langchain-openai For Google models:

pip install langchain-google-genai

Verify the Installation

A helper script check_installation.py validates that the langchain package and required environment variables are correctly set. Run: python check_installation.py The script checks: langchain package is installed.

OpenAI API key is present (if using OpenAI models).

Other environment configurations are functional.

Environment Setup: API Keys

Most LLM providers require an API key, which LangChain reads from environment variables. For OpenAI, set OPENAI_API_KEY either temporarily in the shell: export OPENAI_API_KEY="sk-..." or persistently in a .env file placed at the project root: OPENAI_API_KEY="sk-..." Load the file with python-dotenv:

pip install python-dotenv

from dotenv import load_dotenv
load_dotenv()

Model Usage

The tutorial uses OpenAI‑compatible models, so the langchain-openai package is required. Examples include the DeepSeek and Qwen models. The author uses the DeepSeek public‑cloud model named deepseek-v3; if you use the official DeepSeek model, the name should be deepseek-chat. Replace the name accordingly for Qwen.

Full Verification Script (check_installation.py)

# check_installation.py
import os
import sys
from dotenv import load_dotenv

def check_openai_api_key():
    print("
--- Checking OpenAI API Key ---")
    load_dotenv()
    openai_api_key = os.getenv("OPENAI_API_KEY")
    if not openai_api_key:
        raise ValueError("OPENAI_API_KEY not found in environment variables. Please set it in a .env file.")
    print("  - Found OPENAI_API_KEY environment variable.")
    print(f"    (Key: {openai_api_key[:5]}...{openai_api_key[-4:]})")
    return True

def main():
    """Simple script to verify LangChain and related packages are installed and required env vars are set."""
    print("--- LangChain Installation and Environment Check ---")
    # 1. Verify python-dotenv import
    try:
        from dotenv import load_dotenv
        print("
[SUCCESS] `python-dotenv` imported successfully.")
    except ImportError as e:
        print(f"
[ERROR] `python-dotenv` import failed: {e}")
        print("Please run: pip install python-dotenv")
        return
    # 2. Verify LangChain import
    try:
        import langchain
        from langchain.prompts import PromptTemplate
        from langchain.schema import BaseMessage
        print(f"
[SUCCESS] `langchain` imported successfully. Version: {langchain.__version__}")
    except ImportError as e:
        print(f"
[ERROR] `langchain` import failed: {e}")
        print("Please run: pip install langchain")
        return
    # 3. Verify OpenAI integration package and API key
    try:
        from langchain_openai import OpenAI
        print("[SUCCESS] `langchain-openai` imported successfully.")
        check_openai_api_key()
        api_base = os.getenv("OPENAI_API_BASE")
        if api_base:
            print(f"  - Found OPENAI_API_BASE: {api_base}")
        else:
            print("  - [INFO] OPENAI_API_BASE not set. It's optional for custom endpoints.")
            print("    To use a custom endpoint, add: OPENAI_API_BASE=\"https://your-endpoint.com/v1\" to .env")
    except ImportError as e:
        print(f"
[WARNING] `langchain-openai` not installed or failed to import: {e}")
        print("If you need OpenAI models, run: pip install langchain-openai")
    print("
--- Check Complete ---")
    print("Environment is ready. You can proceed to the next LangChain chapters.")

if __name__ == "__main__":
    main()

References

How to: install LangChain packages – https://python.langchain.com/docs/how_to/install_langchain

How to: use LangChain with different Pydantic versions – https://python.langchain.com/docs/how_to/pydantic_compatibility