# Qdrant — AI Compare Ninja

> Self-host open-source vector search or use it as a managed cloud with inference.

## Overview
- **Category:** Development & Code (Databases, vector search & SQL)
- **Status:** Active · checked - Active open-source version plus Free, Standard, Premium, hybrid, and private cloud offers.
- **Paid from:** pay as you go / Hourly (Cloud Standard)
- **Free access:** Free plan (1 cluster, 0.5 vCPU, 1 GB RAM, 4 GB disk ≈ 1M vectors at 768 dimensions.)
- **Website:** https://qdrant.tech/
- **Canonical URL:** https://www.aicompare.ninja/en/tool/qdrant

## What it does
- Open-source vector database with filtering, hybrid search and REST/gRPC APIs
- Self-host or use as managed cloud with inference
- Vertical and horizontal scaling plus backup and disaster recovery
- Up to 5M free inference tokens per model/month on paid cloud
- Hybrid and private cloud for isolated or air-gapped environments

## In brief
Qdrant is an open-source vector database written in Rust with filtering, hybrid search and REST/gRPC APIs. It can be self-hosted or used as managed, hybrid or private cloud.

Cloud Free includes one cluster with 0.5 vCPU, 1 GB RAM and 4 GB disk. Standard bills resources hourly; Qdrant does not publish one universal monthly entry price.


## Limits
The free cluster is permanently free without a credit card, but suspends after one inactive week and is deleted after four weeks unless reactivated. Qdrant estimates 1 GB RAM and 4 GB disk at roughly one million 768-dimensional vectors. Standard pricing depends on vCPU, RAM, storage, backups, and inference tokens and is exposed through the cloud calculator rather than one fixed entry price.

## Pricing tiers
| Tier | Price | Period | What you get |
| --- | --- | --- | --- |
| Open source | free | Self-hosted | Apache 2.0 licence; REST and gRPC; Own infrastructure cost; No vendor cloud limits; Self-managed operations, backups, and scaling |
| Cloud Free | free | Month | 1 single-node cluster; 0.5 vCPU and 1 GB RAM; 4 GB disk; About 1M vectors at 768 dimensions; Free and external inference models |
| Cloud Standard | pay as you go | Hourly | Dedicated resources; Vertical and horizontal scaling; Backup and disaster recovery; 99.5% uptime SLA in the Standard column of the plan table; Up to 5M free inference tokens per model/month on paid cloud |
| Premium / hybrid / private | contact sales | Contract | SSO and private VPC links; 99.9% uptime SLA on Premium, 99.95% for Multi-AZ setups; Hybrid deployment in your infrastructure; Private cloud for isolated or air-gapped environments; Premium support |

## Sources
All details regarding plans, limits, and features are based on official vendor documentation and the primary sources verified below.

- https://qdrant.tech/pricing/
- https://qdrant.tech/documentation/cloud/create-cluster/
- https://qdrant.tech/cloud-inference/
- https://qdrant.tech/cloud/

---
*Source: AI Compare Ninja (2026) - https://www.aicompare.ninja/en/tool/qdrant*