Closed slavakurilyak closed 9 months ago
Hey there, @slavakurilyak!
I'm going to close this one as a duplicate
so we can consolidate all of the Vector, Memory, RAG, storage, etc. discussions into the first issue that brought them up (#144) and reduce the overall noise in the projects' Issues.
Is your feature request related to a problem? Please describe.
It would be great to see Open Interpreter use retrieval-augmented generation (RAG) to answer questions about source code, docs, or both.
Describe the solution you'd like
I would like Open Interpreter to perform retrieval on existing codebases and/or docs so that code generations are more grounded and the models are hallucinating less.
Describe alternatives you've considered
autogen's Auto Generated Agent Chat
Meta's Retrieval-Augmented Generation (RAG) for Knowledge-Intensive NLP Tasks
Additional context
Related notebook on how to change what OI knows which lets you use RAG to teach OI new things