Vespa is a full-featured text search engine and supports both regular text search and fast approximate vector search (ANN). This makes it easy to create high-performing search applications at any scale, whether you want to use traditional techniques, or a modern vector based approach. You can even combine both approaches efficiently in the same query, something no other engine can do. more
Recommendation, personalization and targeting involves evaluating recommender models over content items to select the best ones. Vespa lets you build applications which does this online, typically combining fast vector search and filtering with evaluation of machine-learned models over the items. This makes it possible to make recommendations specifically for each user or situation, using completely up to date information. more
Providing direct answers to questions is useful when creating chatbots, virtual assistants, and to answer some search queries. Modern question answerers combine evaluation of transformer based language models with efficient retrieval of candidate text snippets. Vespa makes it easy to build question answering systems combining the state of the art in research with performance suitable for production. more
Applications such as e-commerce use a combination of structured data and text and need to provide structured navigation - grouping data dynamically for navigation and filtering - in combination with search and recommendation. Vespa provides all the features required for this with great performance, which makes it possible to realize functionally complete applications leveraging structured data on a unified architecture. more
Combine search in structured data, text and vectors in one query to achieve functionality and performance which is simply impossible with other technologies.
Vespa is engineered around scalable and efficient support for machine-learned model inference, and supports most machine-learned models from most tools.
Vespa will automatically keep data distributed over nodes, and redistribute in the background on changes. No need to worry about how data is divided and distributed.
When Spotify wanted to implement semantic search using vector embeddings, they turned to Vespa for its support for fast ANN search in combination with all the other needs of real applications, such as using ranking functions combining vector similarity with other signals. Read more.
Yahoo runs about 150 Vespa applications, including selecting the personalized content on all content pages in real time and the targeted ads of one of the worlds largest ad exchanges. In total these applications serve close to a billion users, at a rate of 500 000 queries per second.
OkCupid chose Vespa over Elasticsearch for their solution to find matches for users in real time, due to Vespa's automatic data management, flexible ranking, and support for adding new fields for filtering and sorting quickly without refeeding all data. Read more.