framework
LangChain
LangChain is a broad framework for building LLM applications with chains, tools, retrieval, and integrations used alongside agent orchestrators.
- Python
- JavaScript
- framework
Website: python.langchain.com/
GitHub: github.com/langchain-ai/langchain
Languages: Python, JavaScript
License: MIT
Best for
- Composable LLM app building blocks
- RAG pipelines and tool integrations
- Teams standardizing on a large connector ecosystem
Pros and cons
Pros
- Huge integration surface
- Familiar abstractions for many teams
- Works with LangGraph for agent control
Cons / trade-offs
- Abstraction layers can obscure behavior
- Not all apps need the full framework
Overview
LangChain provides building blocks for LLM applications: prompts, tools, retrievers, and integrations. For modern agent control flow, many teams pair it with LangGraph.
Fit signals
Use LangChain when ecosystem connectors and shared abstractions speed you up more than they cost in complexity.
Production note
Prefer thin, explicit modules for critical paths. Framework magic is fine for prototypes; production needs inspectable tool and retrieval boundaries.
Related reading
LangGraph
LangGraph is a library for building stateful, graph-based agent workflows with explicit control flow and durable execution patterns.
How to Choose an Agent Framework
A practical framework selection guide for agentic AI: control model, language, multi-agent needs, ecosystem, and production constraints.