Qdrant

Qdrant is an open-source vector database and semantic search engine written in Rust. It stores vectors together with JSON payloads, then exposes REST APIs for collection management, point upserts, filtering, search, snapshots, and distributed deployment. Retrieval teams can combine dense and sparse vectors, use attribute filters, and model several vectors for one object. The deployment choice is unusually broad: run the open-source engine on your own infrastructure, start with a small managed Cloud cluster, scale dedicated resources, or discuss hybrid and private-cloud arrangements. The free Cloud tier suits testing and prototypes, while production capabilities such as highly available setups, backups, disaster recovery, and stronger uptime commitments sit in paid tiers. Qdrant works well for retrieval-augmented generation, semantic product or document search, and recommendation systems, but adoption still requires capacity planning. Vector data can consume substantial memory, storage-optimized configurations may trade search speed for cost, and heavy writes can compete with query workloads while background optimization runs. Teams should also compare operational responsibility: self-hosting offers control but leaves infrastructure and recovery work to the operator, whereas managed tiers reduce that burden at a usage-based or negotiated cost.