LangChain
Foundational composable library for chains, agents, and memory
View on GitHubWhat it solves
The wiring problem in LLM development: connecting models to tools, data sources, and memory without writing glue code from scratch for every integration.
Overview
LangChain introduced the chain and agent primitives that the modern LLM ecosystem is built on. Core abstractions: chains (composable sequences of LLM calls), tools (functions an LLM can invoke by name), and memory (state that persists across turns). Most major frameworks (LangGraph, LangFlow, DeepAgents) layer on top of it rather than replacing it.
Key facts
- Language
- Python
- License
- MIT
- Maturity
- Community canon
- Maintainer
- langchain-ai
- Reviewed
- 2026-06-11
Where to start
The agents concept doc explains the ReAct loop and tool-calling model that underpins every LangChain agent, read this before reaching for LangGraph.
docs/docs/concepts/agents.mdxTechnologies
More in this domain
LangGraph
Stateful multi-agent orchestration for LLM applications
The state management problem in multi-agent systems: how you build agents that pass context across turns without losing the execution thread.
AutoGen
Multi-agent conversation framework from Microsoft Research
How to coordinate multiple specialized agents toward a shared goal without writing low-level orchestration logic by hand.
CrewAI
Role-based multi-agent orchestration for collaborative task execution
The boilerplate problem in multi-agent setup: defining roles, delegation rules, and task routing without writing a custom orchestration graph.