Vespa

Vespa is a search and recommendation serving engine that combines vector, text and structured retrieval with on-node model inference. It is developed by Vespa.ai through the open-source vespa-engine project, whose public repository dates from 2016, and it is licensed under Apache-2.0, an open-source licence. Buyers can run it self-managed or use Vespa Cloud, the managed offering. The latest release is v8.753.16, published on 16 September 2026. Pricing for the cloud service is not published as a list: the pricing page only leads to a calculator, so costs must be worked out there or confirmed with the vendor. The open-source engine is free to self-host, with cost arising from the infrastructure it runs on. Vespa suits teams that want a single engine to handle several kinds of retrieval, such as combining vector similarity with text search and structured filters, and to run machine-learned models close to the data. It is also categorised as a search engine, so it can be compared with other search products as well as with dedicated vector databases. The project shows as active, with a release in September 2026.