Qdrant Vector Search Engine

Build a Similarity Search Engine on Qdrant Cloud

Store embeddings, pick a distance metric, and get the nearest vectors back in milliseconds with filters applied during the search. Start free on Qdrant Cloud with no credit card.

TRUSTED BY ENTERPRISE TEAMS

Approximate nearest neighbor

Find the Nearest Vectors Fast With Approximate Nearest Neighbor Search

Qdrant indexes vectors in an HNSW graph, a graph-based approximate nearest neighbor index that makes large-scale search fast at a small, measurable recall cost.

An approximate nearest neighbor search walking the HNSW graph from an entry point through hops to the query's nearest neighbors
  • Tune speed against recall per query.

    Raise hnsw_ef to visit more of the graph for higher recall, or lower it for speed.

  • Check against exact results.

    Set exact: true to run a full scan and compare the approximate results with the ground truth.

  • Choose what sits in memory.

    Keep vectors or the index on disk for collections that outgrow RAM, and keep the hot path in memory.

Start free with Qdrant Cloud

Results

What Teams Measured After Moving Similarity Search to Qdrant

Bazaarvoice

about 98% nearest neighbor recall with sub-100 ms queries over billions of vectors on Qdrant Hybrid Cloud.

Read the Bazaarvoice case study

Himalayas

about 30 ms median and 200 ms p95 inside Qdrant across 25M+ requests a month.

Read the Himalayas case study

Pathwork

query time cut from 9 seconds to 2 seconds per query after moving retrieval to Qdrant Cloud.

Read the Pathwork case study
Start free with Qdrant Cloud

Getting started

From Sign-Up to Your First Similarity Query

Create a free cluster.

Sign up for Qdrant Cloud with no credit card.

Create a collection.

Set the vector size and the distance metric your embedding model expects.

Upsert and query.

Upload vectors with their payloads, then send a query vector and get the nearest matches back.

Start Free

Common Questions

How do I get started with Qdrant Cloud?

A free Qdrant Cloud cluster has 1 GB RAM, 0.5 vCPU, 4 GB disk, and one node, enough for about 1 M vectors of 768 dimensions, with no credit card. Move to a Standard cluster when you need dedicated resources, multi-node high availability, backups, and SLAs.

Do I need to normalize vectors for cosine similarity?

Qdrant normalizes vectors when you add them to a collection configured for cosine distance, so you send raw embeddings and the engine handles the rest.

How accurate is approximate nearest neighbor search?

Recall is a setting, measured against a full scan. Raise hnsw_ef for higher recall, or set exact: true on a query to get exact results and see the difference on your data.

Can I filter a vector similarity search engine by metadata?

Yes. Store a JSON payload with each vector, index the fields you filter on, and Qdrant applies the filter during the graph search so the nearest results are the ones that match.

Start Searching by Similarity

Pick a metric, upload your embeddings, and run filtered nearest neighbor queries on a free Qdrant Cloud cluster.

Rocket flying over globe illustration