LightRAG - A simple and fast RAG that beats GraphRAG? (paper explained)

[@pranavghoghari600]

Nicely explained. Keep up the good work.

[@AIBites]

Thank you. May I know if you would be interested in more of theory or hands on implementation style videos?

[@pranavghoghari600]

@AIBites A balance between both. Core theories explained and maybe a quick implementation video. Since the implementation/use case will be different for every viewer but the theory paired with a simple quick example works perfectly.

[@AIBites]

Thanks. That's quite a valuable and sensible feedback

[@karthikreddy9504]

Can we use lightrag to pass the context to a fine tuned LLM?

[@AIWhale3]

What databases are used in light rag? Do you use both a vector and graph db?

[@AIBites]

though it uses graph structures for indexing, I believe in terms of storing the embeddings, its just any other vector DB

[@antonijo01]

Is this method good for complex large codebases?

[@AIBites]

It needs testing to see how well it copes with huge DBs. At least I feel it's going to do better than Naive RAG.

[@antonijo01]

Is this method good for complex large codebases?

[@AIBites]

good question and a nice candiate to test :-) As these techs are evolving fast, I believe only testing will showcase its limitations and potentials