概述
Vespa is an open-source search and AI engine that combines vector search, text search and structured filtering with machine-learned ranking at low latency.
详情
Vespa is an open-source, distributed compute engine for search and AI. It combines vector search, text search, and structured filtering with machine-learned ranking, and serves low-latency queries over large datasets.
This package deploys the official vespaengine/vespa image as a single container running both the config server and the services role. The config server / deployment API is published on port 19071, and the query and document HTTP API on port 8080. Application data and logs are persisted in named volumes mounted at /opt/vespa/var and /opt/vespa/logs. Vespa starts serving the HTTP API on port 8080 only after an application package is deployed to the config server.
Typical use cases include vector and hybrid search, retrieval-augmented generation (RAG) for AI applications, recommendation and ranking systems, and large-scale document search.


