LlamaIndex is an open-source framework for building AI applications that connect language models to external data, tools, and structured workflows. It provides components for ingesting documents, indexing information, retrieval-augmented generation, agents, workflows, and integrations with hundreds of model, embedding, vector-store, and data-source providers. Developers can start with a bundled package or assemble a smaller stack from the core library and selected integrations.

The broader LlamaIndex ecosystem also includes LlamaParse and related document services for OCR, parsing, structured extraction, indexing, and document agents. This makes the framework especially strong for applications centered on private or complex knowledge sources such as reports, files, databases, and enterprise content. LlamaIndex is best suited to developers building custom RAG, search, knowledge, and agentic applications who want a large ecosystem of data connectors and reusable abstractions.