Emad@EMostaque
71AI 编辑部评分,满分 100
2026-05-13 23:38· 90天前
AI 导读

在构建多智能体生产级系统时,PostgreSQL虽可靠但检索速度不足。团队为此开发并开源了psql_bm25s,这是一个原生PostgreSQL访问方法,实现了精确的BM25检索。其在标准基准测试中比pg_search快约23倍,使得检索不再成为性能瓶颈和成本负担,智能体得以高效查询数据,为大规模自主智能体应用铺平道路。

When you want to move from single agent SQLite on something like QMD, PostgreSQL is a great choice for multi agent and production quality, but not as snappy.

So we made it much more snappy with BM25 & open sourced it.

More soon for planetary scale sovereign agents

Intelligent Internetso we built psql_bm25s. exact BM25 retrieval. native Postgres access method. ~23x faster than pg_search on the standard benchmark. retrieval stops being a budge...

来源:Emad · x.com

Emad · @EMostaque · X·2026-05-13 23:38·90天前
AI 导读

在构建多智能体生产级系统时,PostgreSQL虽可靠但检索速度不足。团队为此开发并开源了psql_bm25s,这是一个原生PostgreSQL访问方法,实现了精确的BM25检索。其在标准基准测试中比pg_search快约23倍,使得检索不再成为性能瓶颈和成本负担,智能体得以高效查询数据,为大规模自主智能体应用铺平道路。

When you want to move from single agent SQLite on something like QMD, PostgreSQL is a great choice for multi agent and production quality, but not as snappy.

So we made it much more snappy with BM25 & open sourced it.

More soon for planetary scale sovereign agents

Intelligent Internetso we built psql_bm25s. exact BM25 retrieval. native Postgres access method. ~23x faster than pg_search on the standard benchmark. retrieval stops being a budge...

来源:Emad· x.com