langchain

name	langchain
description	Framework for building LLM-powered applications with agents, chains, and RAG. Supports multiple providers (OpenAI, Anthropic, Google), ReAct agents, tool calling, memory management, and vector store retrieval. Use for building chatbots, question-answering systems, autonomous agents, or RAG applications. Use when this capability is needed.
metadata	{"author":"AcidicSoil"}

LangChain - Build LLM Applications with Agents & RAG

LangChain v1 is the fast way to build provider-agnostic agents and LLM-powered applications. LangChain agents run on top of LangGraph, so you can start high-level with create_agent(...) and drop to LangGraph when you need more explicit control.

When to use LangChain

Use LangChain when you want to:

• build agents quickly with create_agent(...)
• connect to OpenAI, Anthropic, Google, and other providers through dedicated integration packages
• add tools, structured output, and retrieval without hand-writing graph orchestration
• prototype RAG workflows before dropping into LangGraph for more control

Use LangGraph instead when you need:

• explicit stateful workflows with loops, Command, and Send
• persistence, interrupts, or custom orchestration logic as first-class concerns
• deeper control over node boundaries and execution flow

Quick start

Installation

# Core framework
pip install -U langchain

# Provider integrations
pip install -U langchain-anthropic
pip install -U langchain-openai


pip install -U langchain-community langchain-chroma


pip install -U langchain-text-splitters