Cloud & Architecture · Freemium

Qdrant

Qdrant is an open-source vector database and search engine written in Rust, used for semantic search, retrieval-augmented generation (RAG), recommendations and memory for AI agents. Developers can run the open-source version themselves, or use Qdrant Cloud, the company's managed service on AWS, Google Cloud and Microsoft Azure. Qdrant stores vectors with JSON payloads that can be filtered during search, supports hybrid dense and sparse search, and offers quantization to reduce memory use. Qdrant Cloud adds managed clusters with vertical and horizontal scaling, highly available multi-node setups, backups and disaster recovery, shard splitting, GPU indexing on paid tiers, and Cloud Inference for generating embeddings inside the service. Clusters can be managed through an API, Terraform, Pulumi or a CLI. Pricing as of Nov 4, 2026, from Qdrant's official pricing page: the Free Tier is free forever and gives a single-node cluster with 0.5 vCPU, 1 GB of RAM and 4 GB of disk, intended for testing and prototypes. The Standard Tier is usage-based: you pay hourly for the vCPU, memory and storage your clusters use, plus backup storage and any paid inference tokens, with a 99.5% uptime SLA. The Premium Tier requires a minimum spend and adds SSO, private VPC links, a 99.9% uptime SLA (99.95% multi-AZ) and 24/7 support. Qdrant does not publish a single per-query rate, because cost depends on the cluster size you choose; its online calculator estimates a monthly figure. Beyond the managed service, Qdrant offers Hybrid Cloud, which runs Qdrant-managed clusters on a customer's own infrastructure so data stays inside their network, and Private Cloud for fully isolated or air-gapped deployments. Teams can migrate from self-hosted open source to Qdrant Cloud with an official migration tool. Because billing follows provisioned resources rather than reads and writes, Qdrant suits steady, high-volume query workloads where predictable costs matter more than paying only for usage. Metarix covers Qdrant in its comparisons of vector databases for AI agents.