Repolode
Submit
qdrant

qdrant/qdrant-client

Official Python client library for the Qdrant vector search engine, with sync and async REST/gRPC access and a local in-memory mode.

View on GitHubWebsite
Stars
1,373
Forks
357
Open issues
223
Last commit
5 hours ago

What is it?

qdrant-client is the official Python SDK for Qdrant, a vector search engine. It includes type definitions for the Qdrant API and exposes both synchronous and asynchronous requests over REST and gRPC, along with helper methods for common tasks such as bulk uploading a collection.

The client also offers a local mode that runs the same API without a server. Passing ":memory:" or a file path creates an embedded database, which suits prototyping, notebooks and CI test suites. When more capacity is needed, the same code can point at a running Qdrant server or Qdrant Cloud cluster by changing the connection parameters.

An optional Inference API can generate embeddings locally through FastEmbed, with optional GPU support, or remotely through models hosted in Qdrant Cloud, so text or images can be inserted and queried without a separate embedding pipeline.

Key features

  • 01Type hints for all Qdrant API methods
  • 02Synchronous and asynchronous clients over REST and gRPC
  • 03Local in-memory or on-disk mode without a running server
  • 04Inference API with local FastEmbed or remote Qdrant Cloud models
  • 05Helper methods for bulk uploading vectors and points

Use cases

  • Building semantic search or retrieval-augmented generation backends on Qdrant
  • Running vector search tests in CI pipelines without deploying a server
  • Prototyping embedding-based search in Jupyter or Colab notebooks
  • Writing async Python services that query a Qdrant cluster

Quick start

  1. 1

    Install the Python client from PyPI

    pip install qdrant-client
  2. 2

    Start a Qdrant server locally with Docker

    docker run -p 6333:6333 qdrant/qdrant:latest
  3. 3

    Connect to the server from Python using QdrantClient(url="http://localhost:6333")

Things to know

  • Some features such as remote inference are only available on paid Qdrant Cloud plans.
  • Running the server for full use requires Docker or a separate Qdrant server deployment.

Topics

#vector-database#vector-search#python#embeddings#semantic-search#rag#grpc#qdrant#vector-search-engine

FAQ

What is qdrant-client?
Official Python client library for the Qdrant vector search engine, with sync and async REST/gRPC access and a local in-memory mode. qdrant-client is the official Python SDK for Qdrant, a vector search engine. It includes type definitions for the Qdrant API and exposes both synchronous and asynchronous requests over REST and gRPC, along with helper methods for common tasks such as bulk uploading a collection.
Is qdrant-client free to use? What license does it have?
qdrant-client is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is qdrant-client still maintained?
Actively maintained. The last commit to qdrant-client was on Oct 22, 2026.
What can I use qdrant-client for?
Building semantic search or retrieval-augmented generation backends on Qdrant Running vector search tests in CI pipelines without deploying a server Prototyping embedding-based search in Jupyter or Colab notebooks Writing async Python services that query a Qdrant cluster
How do I install qdrant-client?
Install the Python client from PyPI: pip install qdrant-client Start a Qdrant server locally with Docker: docker run -p 6333:6333 qdrant/qdrant:latest Connect to the server from Python using QdrantClient(url="http://localhost:6333")
Where can I get qdrant-client?
The source code and installation instructions are on GitHub: https://github.com/qdrant/qdrant-client

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects