Vespa.ai

Vespa is an AI search platform for querying, organizing and running inference over vectors, tensors, text and structured data. It addresses applications that combine search, retrieval-augmented generation, recommendations, personalization, ad targeting, semi-structured navigation and personal or private search, including workloads with billions of changing data items and high query volumes.

Its capabilities include vector and tensor search, positional text search with configurable linguistics and dynamic snippets, structured search with exact, range, fuzzy and regex operators, and data-aware query planning. Vespa supports distributed machine-learned ranking using ONNX or XGBoost models, multiple rank profiles and multi-phase ranking, plus continuous deployment and automated cluster scaling. Vespa Cloud provides managed operations, security features and deployment on managed AWS and GCP accounts or in customer-owned accounts.

Vespa is aimed at developers and teams building data-driven search and AI applications. The pages describe Vespa Cloud as a fully managed service with automated scaling and continuous platform upgrades, but do not state pricing, editions, trials or seat-based and usage-based packaging.