---
title: "BM25"
description: "2 posts about BM25, summarised, each linking to the original."
---

# BM25
> 2 posts about BM25, summarised, each linking to the original.

## Articles

### [Lakebase Search: State-of-the-art full text and vector search for Postgres](https://yomu.fyi/post/lakebase-search-state-of-the-art-full-text-and-vector-search-for-postg.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Zhou Sun, Jinjing Zhou, Keming Yang, Usamoi Cui, Pranav Aurora
- Published: Sep 28, 2026
- Type: Announcement

Traditional OLTP systems like Postgres face challenges meeting the low-latency, high-accuracy search demands of AI agents, often requiring standalone search engines with ETL pipelines. Lakebase introduces lakebase\_vector and lakebase\_text, two extensions that bring scalable approximate nearest neighbor search and BM25 full-text search directly into Lakebase Postgres. lakebase\_vector overcomes pgvector’s memory-bound HNSW index by using hierarchical IVF clustering and binary quantization, achieving 71-millisecond P99 latency at 97 percent recall on the 100-million-vector LAION benchmark. The extensions decouple storage from compute for pay-per-use scaling, enable fast parallel index builds, and support native hybrid search that fuses vector similarity with keyword relevance in a single SQL query. Early adopters like Conexiom reduced infrastructure costs by three times while achieving five times higher throughput, and the service is generally available on AWS and Azure.


### [Same Query, Same Score: Consistent Ranking with StableTfl in Atlas Search](https://yomu.fyi/post/same-query-same-score-consistent-ranking-with-stabletfl-in-atlas-searc.md)
- Company: [MongoDB](https://yomu.fyi/company/mongodb.md)
- Published: Sep 25, 2026
- Type: Explainer

Paginated search in Atlas Search can show duplicate or skipped results across pages because BM25 scores drift between replicas due to changing corpus statistics. The usual workarounds like sticky routing or index snapshots require extra state and are cumbersome. The solution is StableTfl, a scoring function that uses only term frequency and document length, with term length as a proxy for rarity based on Zipf’s brevity law, eliminating corpus statistics and requiring no extra state. StableTfl is contributed to Apache Lucene and is available as a per-field similarity option in Atlas Search. Benchmarks on BEIR show StableTfl achieves 0.315 NDCG@10 compared to BM25’s 0.346, a 9% relative drop, making it suitable for paginated search, RAG pipelines, and multi-replica deployments where consistency matters more than peak relevance.