LangChain
Framework for building LLM-powered applications with memory, tools, and agents
Visit Product
353 upvotes
3,694 views
About
LangChain is an open-source framework that makes it dramatically easier to build applications powered by large language models. It provides a standard interface for chaining together LLM calls, memory systems, external tools, and data sources — enabling developers to create sophisticated AI applications like chatbots, document Q&A systems, autonomous agents, and more without reinventing the wheel.
At its core, LangChain offers abstractions for the most common LLM application patterns: chains (sequences of LLM calls), agents (LLMs that decide which tools to use), memory (persisting context across conversations), and retrievers (fetching relevant information from external sources). It integrates with virtually every major LLM provider including OpenAI, Anthropic, Google, Hugging Face, and Cohere.
LangChain has become the de facto standard framework for LLM application development, with a massive community, extensive documentation, and integrations with over 200 tools and data sources. It is available in both Python and JavaScript/TypeScript, and LangChain Inc. also offers LangSmith (for debugging and monitoring) and LangGraph (for building stateful multi-agent workflows).
At its core, LangChain offers abstractions for the most common LLM application patterns: chains (sequences of LLM calls), agents (LLMs that decide which tools to use), memory (persisting context across conversations), and retrievers (fetching relevant information from external sources). It integrates with virtually every major LLM provider including OpenAI, Anthropic, Google, Hugging Face, and Cohere.
LangChain has become the de facto standard framework for LLM application development, with a massive community, extensive documentation, and integrations with over 200 tools and data sources. It is available in both Python and JavaScript/TypeScript, and LangChain Inc. also offers LangSmith (for debugging and monitoring) and LangGraph (for building stateful multi-agent workflows).
Product Features
- Chains: composable sequences of LLM prompts and transformations
- Agents: autonomous LLM systems that select and use tools
- Memory: short-term and long-term context for conversational applications
- RAG support: document loading, splitting, embedding, and retrieval
- 200+ integrations with LLM providers, vector stores, and tools
- LangSmith: tracing, testing, and monitoring for LLM apps
- LangGraph: framework for multi-agent and stateful workflows
- Available in Python and JavaScript/TypeScript
- Expression Language (LCEL) for declarative chain composition
- Agents: autonomous LLM systems that select and use tools
- Memory: short-term and long-term context for conversational applications
- RAG support: document loading, splitting, embedding, and retrieval
- 200+ integrations with LLM providers, vector stores, and tools
- LangSmith: tracing, testing, and monitoring for LLM apps
- LangGraph: framework for multi-agent and stateful workflows
- Available in Python and JavaScript/TypeScript
- Expression Language (LCEL) for declarative chain composition
About the Publisher
LangChain was created by Harrison Chase and released in October 2022. The project quickly became one of the fastest-growing open-source repositories in history, amassing tens of thousands of GitHub stars within months. LangChain Inc. was founded in 2023 and raised $25 million in seed funding from Benchmark, followed by a $35 million Series A. The framework has over 2,000 contributors and is used by the majority of companies building LLM applications in production.