Qdrant
Open source vector search engine written in Rust, optimized for performance, available self-hosted or in the cloud
Overview
Qdrant is an open source vector search engine written in Rust, optimized for performance and low latency on large volumes of vectors. The code is licensed under Apache 2.0 and can be self-hosted for free without limits. Qdrant Cloud, the vendor's managed service, offers a permanent free plan with a cluster at 0.5 vCPU, 1 GB RAM, and 4 GB disk for testing and prototypes. The Standard plan, intended for production, is billed by usage on compute, memory, and storage consumed, with no fixed price displayed: Qdrant directs you to an online calculator to estimate costs. The Premium plan, with SSO, private VPC, and 99.9% SLA, requires a minimum commitment whose amount is not published.
Qdrant is not available in French: the Qdrant Cloud interface and documentation are in English only. The vendor, based in Germany, offers hosting in Europe across all three clouds (AWS, GCP, Azure) starting from the Standard plan, which facilitates documented GDPR compliance by data region. The REST and gRPC API is comprehensive, with official integrations to LangChain, LlamaIndex, Haystack, LangGraph, CrewAI, and AutoGen. The main limitation is the absence of published pricing beyond the free plan: it is impossible to compare a precise monthly cost with competitors without using the calculator or requesting a quote, which complicates budgeting for a small organization.
Our verdict
Best for teams wanting a high-performance vector database with a real free plan for prototyping and European hosting available from the paid plan onward. Not for you if you need to know a fixed monthly price before committing: Qdrant Cloud publishes no pricing beyond the free plan, only a usage calculator.