LangChain

LangChain is a composable framework for building applications with LLMs, featuring LangGraph for constructing robust, stateful, cyclic multi-agent architectures.

When to Use

• Agentic Workflows & Multi-Turn Problem Solving: Building autonomous AI agents with LangGraph and tool calling.
• Complex Retrieval-Augmented Generation (RAG): Orchestrating document chunking, embeddings, vector indexing, and reranking.
• Declarative Chain Composition with LCEL: Composing streaming, async, and batchable chains via LangChain Expression Language.
• Enterprise LLM Observability: Tracking token costs, latency, and step-by-step traces with LangSmith.

Quick Start

from langchain_openai import ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser