The short answer
Qdrant Cloud is billed resource-based (vcpu/ram/disk). It has a free tier: 1GB RAM / 4GB disk cluster (~1M vectors), no card. The paid plans we list start at Usage-based (resource billing for vCPU/RAM/disk) (Standard (Managed Cloud)).
Plans and prices
| Plan | Price | What it covers |
|---|---|---|
| Free Tier | $0 (free forever) | 1GB managed cluster: 0.5 vCPU, 1GB RAM, 4GB disk, no credit card; supports ~1M 768-dim vectors; no SLA/HA. Includes free Cloud Inference with selected models. |
| Standard (Managed Cloud) | Usage-based (resource billing for vCPU/RAM/disk) | Dedicated, scalable clusters; 99.5% single-AZ SLA (99.9%/99.95% multi-AZ); daily backups, zero-downtime upgrades on HA, monitoring, RBAC, free inference tokens. |
| Premium | Custom / minimum spend required | 99.9% Uptime SLA, SSO, faster support response, service credits; contact sales. |
| Hybrid Cloud | Custom (pricing on request) | Qdrant-managed control plane running on your own infrastructure/Kubernetes; custom SLA. |
| Private Cloud | Custom (pricing on request) | Fully isolated, dedicated enterprise deployment with custom SLA and data-residency control. |
How Qdrant Cloud bills you
The pricing model is resource-based (vcpu/ram/disk). Rust-built engine with predictable resource-based pricing, a free always-on cluster, and strong filtering. Best suited to: high-performance open-source vector engine.
The pricing page we read
Source: https://qdrant.tech/pricing/. Prices change, so confirm there before you commit.
Compare Qdrant Cloud with alternatives
Other vector database apis: Pinecone, MongoDB Atlas Vector Search, Zilliz Cloud (Milvus), Supabase Vector (pgvector), Weaviate Cloud.
Looking at quality rather than price? See the Qdrant Cloud score on APIbenchmarks. Full entry: Qdrant Cloud pricing.