RAG and Retrieval Systems for LLMs
13 repos
Libraries, benchmarks, and evaluation frameworks for retrieval-augmented generation (RAG)—a pattern that augments large language models with external knowledge retrieval. Repositories here span implementations of RAG systems (like MiniRAG and SearchLM), evaluation datasets and benchmarks (including COIR and OHR-Bench), and tools for measuring retrieval quality. Most projects are Python-based and focus on improving how LLMs access and incorporate retrieved context to generate more accurate, grounded responses.
Python · 12
retrieval-augmented-generation ·7,738
large-language-models ·6,643
rag ·5,773
benchmark ·3,801
datasets ·3,566
llm ·2,455
information-retrieval ·2,174
nlp ·2,037
ai ·1,811
retrieval ·908
