DocsDatabases

Qdrant

A Rust-written vector database known for fast filtered search and a clean REST + gRPC API.

At a glance

License

Apache 2.0

Hosting

Self-host or Qdrant Cloud

Default index

HNSW (per-shard)

Strength

Filterable payloads at speed

Create a collection

import { QdrantClient } from "@qdrant/js-client-rest";

const client = new QdrantClient({ url: "http://localhost:6333" });

await client.createCollection("docs", {
  vectors: { size: 1536, distance: "Cosine" },
});

Filtered search

Qdrant indexes payload fields alongside vectors, so a query like "top 5 nearest, but only docs where tenant_id = 42" runs inside the HNSW traversal instead of a post-filter. That's a big quality win when your tenants are small.

await client.search("docs", {
  vector: queryVector,
  limit: 5,
  filter: {
    must: [
      { key: "tenant_id", match: { value: 42 } },
      { key: "year",      range: { gte: 2024 } },
    ],
  },
});