langchain-ai/langchain

Health v2 · maintenance only

Adoption not part of the score

145076 stars · 24198 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

LangChain is an open-source Python framework for building LLM-powered agents and applications, providing a configurable agent harness (create_agent) plus 1000+ provider integrations for chat models, embeddings, tools, and vector stores. It anchors an ecosystem including LangGraph for low-level orchestration, Deep Agents for batteries-included agents, and LangSmith for tracing and evaluation.

Use cases

• build an ai agent with tool calling in python
• swap between openai, anthropic, and gemini models without changing code
• build a rag chatbot over my documents
• create a multi-agent workflow with planning and subagents
• chain llm calls with custom tools and middleware
• prototype an llm application with vector store integrations
• add tracing and evaluation to my agent app

When to choose

• you want a model-agnostic framework with a huge integration ecosystem
• you need a customizable agent harness with middleware and tool support
• you want to build RAG pipelines or chatbots quickly in Python
• you plan to scale from simple agents to complex LangGraph workflows

When to avoid

• you need only raw model API calls without abstractions - use provider SDKs directly
• you want a minimal dependency-free solution for simple LLM usage
• you need a JS/TS-first codebase - use LangChain.js instead
• you require strict control over every prompt token and dislike framework magic

Facets

framework · maturity stable

agent-framework llm-inference rag chatbot sdk middleware artificial-intelligence large-language-models chatbots developer-tools python cross-platform llm agents multi-agent model-agnostic integrations langgraph deep-agents openai anthropic gemini ai-agents retrieval-augmented-generation

Member repositories

For agents

markdown · JSON · MCP: product_card(name="langchain-ai/langchain")