mkurman/qdrant

Overview

Qdrant is a high-performance vector similarity search engine supporting dense and sparse vectors, payload indexing and filtering, scalar/PQ quantization, multi-tenancy, and horizontal scaling via clustering. REST and gRPC APIs with async support.

Installation

bash
docker run -p 6333:6333 qdrant/qdrant

Python Client

python
from qdrant_client import QdrantClient, models
import numpy as np

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

client.upsert("documents", points=[
    models.PointStruct(id=1, vector=np.random.rand(384).tolist(), payload={"text": "Paris is capital of France"}),
    models.PointStruct(id=2, vector=np.random.rand(384).tolist(), payload={"text": "Berlin is capital of Germany"}),
])

results = client.search("documents", query_vector=np.random.rand(384).tolist(), limit=5)
for hit in results:
    print(hit.payload["text"], hit.score)

References

• Qdrant docs
• Qdrant GitHub