Qdrant: vector similarity search engine and vector database

Qdrant is a vector similarity search engine and vector database.

Verified

Core facts

Read by a script from the official API, not rewritten by a model.

License
Apache-2.0
Languages
Rust
Maintainer
qdrant
Verified

What it is

Qdrant is a vector similarity search engine and vector database. Qdrant provides a service with an API to store, search, and manage points, which are vectors with an additional payload. Qdrant is tailored for extended filtering support. Qdrant is also available as a fully managed Qdrant Cloud, including a free tier.

Main features

  1. Qdrant supports dense vectors for semantic similarity, sparse vectors for full-text search, and multivector search for objects with multiple embeddings or late interaction models like ColBERT.

  2. A JSON payload can be attached to vectors and filtered with conditions such as keyword matching, full-text, numeric ranges and geo-locations.

  3. Hybrid search combines multiple vectors in a single query, with results merged via configurable fusion strategies such as RRF and DBSF.

  4. The project says distributed deployment scales horizontally with sharding and replication, and can update or resize collections with zero downtime.

  5. Faceting aggregates search results by payload values.

  6. Recommendation uses positive and negative examples to find similar points.

Getting started

  1. The Docker command in this section starts an insecure deployment without authentication, open to all network interfaces.

  2. Qdrant provides a collection of agent skills that bring its vector search capabilities into an AI coding assistant.

  3. The project says to read its installation and security guides before deploying Qdrant to production.

  4. To get started with Qdrant Edge from Python or Rust, initialize an instance of EdgeShard, which exposes methods to manage data, query it, and restore snapshots.

  5. Qdrant provides a REST API with an OpenAPI 3.0 specification, enabling client generation.

  6. Qdrant also provides a gRPC interface.

From the README · bash
docker run -p 6333:6333 qdrant/qdrant
From the README · python
from qdrant_client import QdrantClient

client = QdrantClient(url="http://localhost:6333")
From the README · python
from qdrant_edge import Distance, EdgeConfig, EdgeVectorParams, EdgeShard, Point, UpdateOperation

shard = EdgeShard.create("./shard", EdgeConfig(
    vectors={"my-vector": EdgeVectorParams(size=4, distance=Distance.Cosine)}
))
shard.update(UpdateOperation.upsert_points([
    Point(id=1, vector={"my-vector": [0.1, 0.2, 0.3, 0.4]}, payload={"color": "red"})
]))

Who it is for

The project says Qdrant is written in Rust, which makes it fast and reliable even under high load. The project describes Qdrant Edge as a lightweight version of Qdrant designed for edge devices and resource-constrained environments. Unlike Qdrant Server, which uses a client-server architecture, Qdrant Edge runs inside the application process.

Discovery constrains search to a specific region of the vector space. Search Relevance Tuning offers tools for adjusting search results, such as MMR and the Relevance Feedback Query. Query Planning and Payload Indexes leverage stored payload information to optimize query execution strategy.

GPU Support is for accelerated indexing, with support for NVIDIA and AMD GPUs. The project says Write-Ahead Logging ensures data persistence with update confirmation, even during power outages. Multitenancy is partitioning of data for multi-user environments.

Observability covers metrics, telemetry, and audit logging for monitoring and debugging. Web UI provides a visual way to interact with data and monitor the health of a deployment. Web UI enables exploring collections, managing data, and interacting with the REST API.

Qdrant connects to embedding providers, AI application frameworks, and data pipeline tools, as well as observability platforms for monitoring and tracing vector search in production. Qdrant is licensed under the Apache License, Version 2.0.

Sources

This page is written from the official sources below. Open them to check the original text.

Verified ·

This page is written by AI from the official sources above and checked automatically. This site has not tested it. Found a mistake? Tell us.