VectorChord-BM25: Supercharging PostgreSQL Full-Text Search

2025-03-03
VectorChord-BM25: Supercharging PostgreSQL Full-Text Search

VectorChord-BM25 is a new PostgreSQL extension leveraging the BM25 algorithm and Block WeakAnd algorithm to significantly improve the speed and accuracy of PostgreSQL's full-text search. It simplifies the search process and seamlessly integrates with PostgreSQL. Compared to ElasticSearch, VectorChord-BM25 achieves 3x higher Queries Per Second (QPS) on average for Top 1000 queries and comparable or even superior NDCG@10 scores, but requires careful alignment of tokenization strategies for fair benchmarking.

Read more
Development

Supercharging Vector Search with ColBERT Reranking in PostgreSQL

2025-01-24
Supercharging Vector Search with ColBERT Reranking in PostgreSQL

Traditional vector search relies on sentence embeddings, potentially losing fine-grained details. ColBERT overcomes this by representing text as token-level multi-vectors, retaining nuanced information and improving accuracy. However, token-level interaction is computationally expensive. This blog post demonstrates combining sentence-level vector search with ColBERT token-level reranking using the PostgreSQL extensions VectorChord and pgvector. This approach performs a fast initial search using sentence embeddings, followed by reranking with ColBERT for improved results. Significant improvements were observed on several BEIR datasets.

Read more
Development vector search