Home / Companies / LangChain / Blog / Post Details
Content Deep Dive

Retrieval

Blog post from LangChain

Post Details
Company
Date Published
Author
-
Word Count
1,028
Language
English
Hacker News Points
212
Summary

LangChain has introduced adjustments to its abstractions to make it easier for other retrieval methods besides the VectorDB object to be used in LangChain. The changes aim to allow retrievers constructed elsewhere to be used more easily in LangChain and encourage experimentation with alternative retrieval methods, such as hybrid search. This is done while maintaining backwards compatibility so that existing chains continue to work as before. However, it's recommended to update from VectorDB chains to the new Retrieval chains as soon as possible since those will be the ones most fully supported going forward.