LlamaIndex

LlamaIndex (formerly GPT Index) is a data framework designed for building LLM applications that work with custom data sources. It specializes in connecting LLMs to external data through indexing, retrieval, and query interfaces.

Key components include: Data Connectors (loading data from various sources), Indexes (structuring data for efficient retrieval), Query Engines (interfaces for asking questions), and various retrieval strategies optimized for different use cases. LlamaIndex excels at RAG applications and knowledge-based assistants.

Compared to LangChain, LlamaIndex is more focused on data retrieval and indexing rather than general LLM application orchestration. Many engineers use both: LlamaIndex for sophisticated RAG implementations and LangChain for agent workflows and chain composition. Understanding both tools helps engineers choose the right approach for each project.