Hybrid search vs. Token pooling benchmarking

Hybrid search vs. Token pooling benchmarking(blog.colivara.com)

3 points by jonathan-adly 1 year ago | 1 comment

Hi HN!

We benchmarked two ways to improve the latency in RAG workflows with a multi-vector setup. Hybrid search using Postgres native capabilities and a relatively new method of token pooling. Token pooling unlocked up to 70% faster latency with <1% performance cost.

jonathan-adly 1 year ago |

Hi HN!

We benchmarked two ways to improve the latency in RAG workflows with a multi-vector setup. Hybrid search using Postgres native capabilities and a relatively new method of token pooling. Token pooling unlocked up to 70% faster latency with <1% performance cost.