RAG & memory · Python
A data framework for connecting private documents to LLMs, with strong RAG and indexing primitives.
Search prompts, categories, platforms and repositories
Ranked list · 3 repositories
The most useful open-source repositories for retrieval-augmented generation, document parsing, vector search and agent memory.
Retrieval quality decides whether a RAG system feels useful or merely plausible. The stack has four parts: ingestion and parsing, chunking and indexing, retrieval and reranking, and answer synthesis with citations. The repositories below specialise in different layers, and many teams combine them rather than choosing one.
Start with a small, messy document set that represents your real content. Measure retrieval hit rate before you tune the generation prompt. If the source chunks are wrong, no prompt will save the answer. Store citations from day one so users can audit the result.
RAG & memory · Python
A data framework for connecting private documents to LLMs, with strong RAG and indexing primitives.
RAG & memory · Python
A RAG engine with document parsing, chunking, retrieval and grounded citations.
RAG & memory · Python
A memory layer for AI assistants that stores and retrieves user preferences across sessions.
Rankings combine stars, recent commit activity, licence clarity and documentation quality. Stars alone reward old projects with large historical audiences, so we weight maintenance and documentation too. We never accept payment for placement; repository pages are written by our editors and reviewed quarterly.
Back to the directory