Qdrant

Platform

Open-source vector search engine, self-hosted or on Qdrant Cloud

Price: Free tier, then paid
Access: API key (optional when self-hosted)

About

Vector database written in Rust for dense, sparse and multi-vector search with payload filtering. Run the Apache-2.0 engine yourself with Docker, or use Qdrant Cloud, which has a free 1 GB cluster and optional hosted embedding inference.

What you can do with it

Store embeddings and retrieve the nearest matches for RAG and semantic search
Combine dense and sparse (BM25) vectors in one hybrid search query
Filter vector results by metadata such as user, date or category

Get started

Start Qdrant with Docker or create a free Qdrant Cloud cluster
Create a collection with a vector size and distance metric
Upsert points and query the nearest neighbours

Example

docker run -d -p 6333:6333 qdrant/qdrant
curl -X PUT http://localhost:6333/collections/test_collection \
  -H 'Content-Type: application/json' \
  --data-raw '{"vectors": {"size": 4, "distance": "Dot"}}'

Details

Hosting: Hosted service, Self-hosted
Available in: Worldwide
Official SDKs: Python, JavaScript/TypeScript, Rust, Go, C#, Java
MCP server: Local