Vespa's May 2026 newsletter covers a broad set of product updates across Vespa Cloud and the open-source engine. Highlights include detailed metric dashboards now exposed to customers, automated index snapshot backups for disaster recovery, fine-grained maintenance window controls, and native embedding integrations with Voyage AI, OpenAI, and Mistral AI. Engine-level improvements include a new text() query operator for raw text matching, cluster-size-independent relevance effort configuration using 'total-' prefixed parameters, boolean array fields for memory-efficient metadata storage, array element-specific matching, search group pinning for consistent pagination, in-memory document IDs for fast retrieval, near-matching-aware ranking, a firstPhaseMax rank feature for relative score filtering, and geo distance grouping. The newsletter also announces the first in-person Vespa community meetup (Vespa Live), new agent skills for Vespa development, and links to recent blog posts and upcoming events.

12m read timeFrom blog.vespa.ai
Post cover image
Table of contents
Calling All Vespians: Vespa Live Has LandedProduct updatesVespa Cloud: Detailed metric dashboardsVespa Cloud: Index backupVespa Cloud: Fine-Grained maintenance controlsVespa Cloud: Voyage AI, OpenAI, and Mistral AI embedding integrationVespa Cloud: Custom resource tagsVespa skills for agentsA new query operator for text matchingCluster-size independent configuration of relevance effortBoolean array fieldsMatch specific array elementsSearch group pinningIn-memory document idsNear matching aware rankingDetect ignored write operationsAccessing the max first phase score in re-rankingGeo distance in groupingImage Search demoVespa LearnWhat’s New on YouTubeBlogs and ebooksUpcoming events

Sort: