Glossary
LlamaIndex
An open-source RAG framework focused on connecting LLMs to external data, with strong document-ingestion tooling and a smaller surface area than LangChain.
The RAGretrieval-memoryA pattern where a model retrieves relevant documents from an external store at query time and conditions its answer on them, instead of relying only on parametric knowledge. Open full entry -focused alternative to LangChainagentsThe earliest widely-adopted LLM agent and RAG orchestration framework (2022), now with the LangGraph extension for stateful multi-step agent workflows. Open full entry . LlamaIndex (originally GPT Index, renamed in 2023) specializes in data ingestion, indexing, and retrieval for LLM applications. Its API surface is smaller than LangChainagentsThe earliest widely-adopted LLM agent and RAG orchestration framework (2022), now with the LangGraph extension for stateful multi-step agent workflows. Open full entry ’s and its agent abstractions are thinner, by design: the framework targets “give an LLM access to your documents” cleanly without taking on the full agent-orchestration domain.
LlamaIndex ships document loaders for hundreds of source types, a flexible chunkingretrieval-memorySplitting source documents into smaller passages for embedding and retrieval, where the chunk size and overlap directly affect retrieval quality and context efficiency. Open full entry and embeddingretrieval-memoryA fixed-size vector representation of a piece of text learned so semantically similar texts land near each other in the vector space, the basis for vector search and most RAG. Open full entry pipeline, vector-store and hybrid-search integrations, and a query engine that handles multi-step retrieval and synthesis. The framework is permissively licensed.
Full coverage at /projects/llama-index.