Qdrant is an open-source vector database and search engine, written in Rust, that stores embeddings with extra data and finds the closest matches with strong filtering.
In plain words
Qdrant stores points: a vector plus a payload, which is extra information such as a price or category. You search for vectors close to a query and can filter on payload fields at the same time. Besides dense vectors that capture meaning, it supports keyword search for exact matches, and combines the two in hybrid search. It runs as a server with official clients for several languages, and there is also a managed cloud version.
A quick example
Imagine an online shop where each product has an embedding of its description and a payload with price and colour. A shopper searches for a warm winter jacket under a set price. Qdrant finds the nearest product vectors and keeps only those whose payload matches the price and colour filters.
Where you'll meet it
- Semantic and hybrid search.
- Retrieval for RAG apps.
- Recommendation and matching.
Related
Sources
- Docsqdrant/qdrant, Qdrant (GitHub) · read 29 Sept 2026
- DocsOverview, Qdrant documentation · read 22 Oct 2026
Checked by WhatIsAI against the sources above · 22 Oct 2026. Spotted an error? Report it.