Vespa

Vespa is an open-source engine for real-time search, recommendation, and data serving at scale. It is designed to handle large, constantly changing datasets while providing low-latency query and update performance. Vespa combines full-text search, vector search, and structured data filtering in a single platform, enabling complex relevance ranking and personalized experiences. It supports approximate nearest neighbor (ANN) search for embeddings, making it suitable for AI-driven applications such as semantic search, recommendation systems, and content discovery.

Core capabilities include flexible schema definition, powerful query language, and built-in support for ranking expressions that combine textual, vector, and metadata signals. Vespa handles data ingestion, indexing, and querying in a distributed architecture, with automatic sharding and replication for high availability and scalability. It also provides features like tensor operations, custom ranking models, and integration with machine learning frameworks for serving trained models in production.