Vespa
Vector databasesAn open-source engine for large-scale search, recommendation, and vector workloads.
Vespa is a mature, big-data serving engine that combines vector search, keyword search, structured filtering, and machine-learned ranking — and applies them over very large datasets with low latency.
It is the heavyweight option: powerful and battle-tested at scale, but a substantial system to learn and operate.
Where it's ideally used
A fit for large-scale search and recommendation where vectors, keywords, and learned ranking must all work together.
Where it doesn't fit
A large, complex platform — far more than a straightforward RAG project needs.