News for hackers. Stuff that matters.

A comment on Myscaledb: Open-source SQL vector database to build AI apps using SQL. Parent comment

Re:Myscaledb: Open-source SQL vector database to build AI ap... by basil-rash on Tuesday April 02, 2024 @04:53AM (#39903617)
This page has a fairly helpful breakdown of the various types of indexes available, but from what I can tell it doesn’t explicitly mention the correctness trade offs made when using an index vs a pure table scan. I know in pgvector HNSW for instance is inexact. But it doesn’t really matter in a RAG context - even the embeddings themselves are stochastic.

https://myscale.com/blog/everything-about-vector-indexing/

[ Parent | on HN ]
Re:Myscaledb: Open-source SQL vector database to build AI ap... by nrjames on Tuesday April 02, 2024 @05:52AM (#39903929)
I'm not sure how much of the underlying Clickhouse feature set is available, but the Clickhouse documentation about index types is thorough and informative.

https://clickhouse.com/docs/en/optimize/sparse-primary-index...

[ Parent | on HN ]