Composable by design
LangChain’s standard interface lets you experiment with different providers, tools, and databases – creating DevEx parity when gaps exist.
One interface,
integrate any LLM.
LangChain is the open, composable framework that provides a standard interface for every model, tool, and database – so you can build LLM apps that adapt as fast as the ecosystem evolves.
Why use LangChain?
Real-time data augmentation
Connect LLMs to live data sources, APIs, and internal systems. LangChain’s vast library of integrations – including model providers, tools, and vector stores – brings your data directly into your AI workflows.
Open and neutral
Evolve your stack without vendor lock-in. LangChain is open source with 600+ integrations. Future proof your app by swapping in components without rewriting your entire application.
Instantly connect to your preferred LLM.
Building agents? Try LangGraph.
LangChain handles integrations across models, tools, and data. LangGraph lets you orchestrate reliable agents with full control.
LangChain FAQs
Is LangChain open source?
Yes - LangChain is an MIT-licensed open-source library and is free to use.
What are the most common ways people use LangChain?
There are many different use cases for LangChain. Some common ones that we see include: chatbots and conversational interfaces, document Q&A and knowledge retrieval systems, and data extraction. LangChain excels when you need to connect LLMs to external data sources, APIs, or tools– anywhere you need maximum integration flexibility.
How do I use LangChain with other products like LangSmith, LangGraph, or LangGraph Platform?
LangChain provides a standard interface for connecting models, tools, and data, then integrates seamlessly with any of the Lang- family products.
Use LangSmith to debug and evaluate your applications with trace-level visibility and monitoring. When building a complex agentic system, use LangGraph for controllable orchestration. To deploy your agents at scale, use LangGraph Platform.
How is LangChain different from LangGraph?
LangChain is designed for connecting LLMs to data sources with minimal setup. LangGraph is our controllable agent orchestration framework, with out-of-the-box state management and human-in-the-loop capabilities. Use LangChain when you need fast integration and experimentation; use LangGraph when you need to build agents that can reliably handle complex tasks.
Can I use LangChain in production?
Yes. LangChain 0.1 and later are production ready (link to docs).We're also committed to no breaking changes on any minor version of LangChain after 0.1, so you can upgrade your patch versions (e.g., 0.2.x) on any minor version without impact. Companies like Rakuten, Cisco, and Moody’s use LangChain in production for business-critical workflows.
Should I start with LangChain or LangGraph for building agents?
Use LangChain for composability and model flexibility— great for quickly chaining LLMs with tools, retrievers, and external data sources. If you're experimenting with different models, prompts, or RAG pipelines, LangChain gives you the building blocks to move fast.
Use LangGraph if you're building an agent, especially one that requires multi-step reasoning, memory, or low-level control. LangGraph allows you to orchestrate complex agent workflows – with built-in statefulness, human-in-the-loop, and first-class streaming.
The biggest developer community in GenAI
1 M+
developers use LangChain
100k+
GitHub stars
#1
Downloaded agent framework
600+
Integrations
Ready to start shipping reliable agents faster?
Get started with tools from the LangChain product suite for every step of the agent development lifecycle.