ConfigsMar 31, 2026·2 min read

Qdrant — High-Performance Vector Database

Vector database and search engine for AI applications. Handles billion-scale similarity search with filtering, sparse vectors, and multi-tenancy. Rust-powered. 30K+ stars.

TO
TokRepo精选 · Community
Quick Use

Use it first, then decide how deep to go

This block should tell both the user and the agent what to copy, install, and apply first.

# Run with Docker
docker run -p 6333:6333 qdrant/qdrant

# Or install Python client
pip install qdrant-client
from qdrant_client import QdrantClient
from qdrant_client.models import Distance, VectorParams, PointStruct

client = QdrantClient("localhost", port=6333)
client.create_collection("docs", vectors_config=VectorParams(size=384, distance=Distance.COSINE))

client.upsert("docs", points=[
    PointStruct(id=1, vector=[0.1]*384, payload={"title": "AI Guide"}),
])

results = client.query_points("docs", query=[0.1]*384, limit=5)
Intro

Qdrant is a high-performance vector database and similarity search engine built in Rust. Designed for production AI applications — RAG pipelines, semantic search, recommendation systems, and anomaly detection. Supports dense and sparse vectors, payload filtering, multi-tenancy, and horizontal scaling. Available as self-hosted (Docker/binary) or managed cloud. 30,000+ GitHub stars.

Best for: Production RAG, semantic search, recommendation engines, and any app needing vector similarity Works with: LangChain, LlamaIndex, Haystack, OpenAI, Cohere, any embedding model

Key Features

Performance

  • Rust-powered — written in Rust for maximum speed and memory safety
  • Billion-scale — handles billions of vectors with quantization and sharding
  • HNSW indexing — fast approximate nearest neighbor search

Advanced Search

  • Hybrid search — combine dense vectors with sparse vectors (BM25-style)
  • Payload filtering — filter results by metadata alongside vector similarity
  • Multi-vector — store multiple vectors per point (e.g., title + content embeddings)

Production Features

  • Multi-tenancy — isolate data per customer within one collection
  • Snapshots — backup and restore collections
  • Horizontal scaling — distributed mode with sharding and replication
  • REST + gRPC APIs

Integrations

LangChain, LlamaIndex, Haystack, Spring AI, AutoGen, CrewAI, and 50+ frameworks.

FAQ

Q: What is Qdrant? A: A Rust-powered vector database for AI applications. Handles billion-scale similarity search with filtering, sparse vectors, and multi-tenancy. 30K+ GitHub stars.

Q: Is Qdrant free? A: Open-source under Apache 2.0 for self-hosting. Qdrant Cloud offers a managed service with a free tier (1GB).

🙏

Source & Thanks

Created by Qdrant. Licensed under Apache 2.0. qdrant/qdrant — 30,000+ GitHub stars

Related Assets

Claude Engineer — Interactive AI Dev CLI

Interactive CLI powered by Claude for software development. Self-generating tools, file operations, web search, code analysis, and project management. 11K+ stars.

TokRepo精选

Phidata — Build & Deploy AI Agents at Scale

Framework for building, running, and managing AI agents at scale. Memory, knowledge, tools, reasoning, and team workflows. Monitoring dashboard included. 39K+ stars.

TokRepo精选

LobeChat — Modern AI Chat Framework & Agent Hub

Open-source AI chat framework with multi-agent collaboration, plugin marketplace, TTS, vision, and file upload. Supports 70+ model providers. Self-hostable. 75K+ stars.

TokRepo精选
◈Home🏆Trending👤Me