Basics of NLP and RAG · Lesson 5
RAG using LangChain and LlamaIndex
LangChain and LlamaIndex are frameworks that package RAG building blocks. They provide document loaders for PDFs, web pages and databases, text splitters, embedding and vector store integrations, retrievers and chains.
LangChain is general-purpose: you compose components (prompt | model | parser) using its expression language and connect to many tools. LlamaIndex focuses on data: indexing, query engines, and advanced retrieval over documents.
Frameworks speed up prototypes, but understand what happens underneath so you can debug and optimise.
Key points
- Both offer loaders, splitters, retrievers and integrations
- LangChain: flexible chains and agents
- LlamaIndex: data indexing and query engines
- Know the internals to debug well