LlamaIndex
Data framework focused on ingesting, indexing and querying private knowledge.
Обзор
LlamaIndex is a data framework that connects language models to private or domain-specific knowledge by handling ingestion, indexing, retrieval and query composition. It solves the problem that the hard part of a retrieval application is the data pipeline, not the model call. It exists because retrieval quality depends on how documents are parsed, chunked and indexed, and those decisions are worth making once in a reusable layer rather than per project. It overlaps with LangChain but leans further toward indexing and query strategy.
Граф технологий
Как LlamaIndex связана с остальным стеком.
Применяется к
- RAG technique Techniques LlamaIndex used for RAG High ✓ source
Инструменты
ПО, которое реализует LlamaIndex или интегрируется с ним.