Qdrant review

Q: How do I install Qdrant?

Run docker run -p 6333:6333 qdrant/qdrant. It needs Node.js >=22.0.0. The install section above has the other options from the project's README.

Qdrant is an open-source vector similarity search engine and vector database written in Rust, with filtering on JSON payloads.

Updated Nov 1, 2026. Review written Oct 22, 2026 from the project's own sources; usage data refreshed every Monday.

TL;DR

Qdrant is a vector database and search engine that stores vectors with JSON payloads and searches them through REST and gRPC APIs. It is for developers building semantic search, recommendation, or RAG retrieval. It supports dense, sparse, and multivector search, hybrid search, and filters applied during HNSW traversal.

35.4kGitHub stars
+129new stars, week of Oct 25
~4.5Mdownloads a week, +5% vs 4-week avg
#19in RAG, data and memory this week

Qdrant at a glance

Category: RAG, data and memory
Maintained by: Qdrant (organization, @qdrant on GitHub)
License: Apache 2.0
Language: Rust, Python, Shell
Repository: github.com/qdrant/qdrant
Website: qdrant.tech
npm package: @qdrant/js-client-rest 1.19.0
PyPI package: qdrant-client 1.19.1
Latest release: v1.19.2 on GitHub, Oct 5, 2026
First published: May 30, 2020

What Qdrant does

Qdrant stores, searches, and manages points, which are vectors with an attached JSON payload.
Qdrant supports dense vectors for semantic similarity, sparse vectors for full-text search, and multivector search for late interaction models like ColBERT.
Qdrant filters on payloads using keyword, full-text, numeric range, and geo conditions combined with should, must, and must_not clauses.
Qdrant combines multiple vectors in one hybrid query and merges results with fusion strategies such as RRF and DBSF.
Qdrant offers built-in vector quantization and on-disk storage to reduce RAM usage and tune speed against precision.
Qdrant scales horizontally with sharding and replication, and supports multitenancy for multi-user environments.
Qdrant provides recommendation from positive and negative examples, discovery search, faceting, and MMR-based relevance tuning.

How it works

Qdrant runs as a server, for example in a Docker container on port 6333. Applications connect through a REST API with an OpenAPI 3.0 spec, a gRPC interface, or official clients for Python, JavaScript/TypeScript, Go, Rust, .NET, and Java. A Web UI lets you explore collections. Qdrant Edge runs inside the application process instead and can sync data with a Qdrant server.

How to install and start using Qdrant

With Docker

docker run -p 6333:6333 qdrant/qdrant

Requirements

The Python client qdrant-client requires Python 3.10 or later.
The JavaScript REST client @qdrant/js-client-rest requires Node.js 22.0.0 or later.
Docker is needed for the README's local quick start.

First steps

Start a local Qdrant server with the Docker command, which exposes port 6333.
Connect from Python with QdrantClient(url="http://localhost:6333") from the qdrant_client package.
Read the installation and security guides before deploying, because the default container runs without authentication.
Follow the Quick Start Guide or the beginner tutorial to build a semantic search engine.

Who Qdrant is for

Developers building semantic search, image search, or recommendation systems on top of embeddings.
Teams building RAG retrieval or AI agent memory that need metadata filtering.
Applications needing hybrid keyword and vector search in a single query.
Edge or offline applications that need in-process vector search through Qdrant Edge.

Limitations to know

Qdrant's default Docker command starts an insecure deployment with no authentication, open to all network interfaces.
Qdrant Edge is labeled Beta on the homepage.
Qdrant Edge's getting-started path covers Python and Rust only.
Qdrant Serverless is listed as coming soon and is not yet available.

Is Qdrant growing? Usage data

In the week of Oct 25–31, 2026, Qdrant ranked #178 on the Devtoolwatch Board and #19 in RAG, data and memory, with a momentum score of 47 out of 100. Its main package had about 4.5M downloads that week, +5% against its own 4-week average.

New GitHub stars per week for the last 12 weeks, from GitHub's stargazer history (today's stargazers by the week they starred).

Show the weekly numbers

New GitHub stars per week for Qdrant
Week of	New stars
Aug 9, 2026	138
Aug 16, 2026	150
Aug 23, 2026	140
Aug 30, 2026	149
Sep 6, 2026	124
Sep 13, 2026	184
Sep 20, 2026	155
Sep 27, 2026	102
Oct 4, 2026	115
Oct 11, 2026	119
Oct 18, 2026	122
Oct 25, 2026	129

On Hacker News

We were not accepted into Google Summer of Code. So, we started our own, 146 points and 46 comments (Feb 24, 2024)
Qdrant, the Vector Search Database, raised $28M in a Series A round, 131 points and 167 comments (Jan 23, 2024)
Serverless Semantic Search, Free tier only, 91 points and 61 comments (Jul 12, 2023)
Qdrant 1.7.0, 87 points and 29 comments (Dec 12, 2023)
Similarity Learning lacks a framework. So we built one, 59 points and 24 comments (Jul 13, 2022)

The Board ranks tools by four signals each week: new stars, new stars for the repo's size, added downloads and download growth beyond the market. How the ranking works.

Qdrant alternatives

Other RAG, data and memory tools on the Board in the week of Oct 25–31, 2026, by momentum:

RAG, data and memory ranked by momentum, the week of Oct 25–31, 2026
Rank	Tool	Stars	Downloads/wk
#1	XERJXERJ is a local-first search engine and MCP server for AI agents, packaged as one Rust binary with an Elasticsearch-compatible API.	17.9k	–
#2	MemvidMemvid is a single-file memory layer for AI agents that packages data, embeddings, search indexes, and metadata into one portable .mv2 file.	17.3k	–
#3	LEANNLEANN is a Python vector database and RAG library that recomputes embeddings on demand to keep indexes small on personal devices.	13.7k	–
#4	GraphifyGraphify is an AI coding assistant skill that turns a codebase, with its docs, PDFs, images and videos, into a queryable knowledge graph.	140k	480.7k
#5	MarkItDownMarkItDown is a lightweight Python utility from Microsoft that converts PDFs, Office documents, images, audio and other files to…	193.2k	4.7M
#6	FirecrawlFirecrawl is an API and open source project to search, scrape, crawl, and interact with web pages, returning LLM-ready markdown or JSON…	197.4k	2.9M

All RAG, data and memory tools, ranked

Recent Qdrant releases

v1.19.2, Oct 5, 2026
v1.19.1, Sep 4, 2026
v1.19.0, Aug 5, 2026
v1.18.3, Jul 17, 2026
v1.18.2, Jun 4, 2026

Qdrant FAQ

Is Qdrant free and open source?

Yes. Qdrant's source code is on GitHub (qdrant/qdrant) under the Apache 2.0 license, so you can use it for free.

How do I install Qdrant?

Run docker run -p 6333:6333 qdrant/qdrant. It needs Node.js >=22.0.0. The install section above has the other options from the project's README.

Is Qdrant still maintained?

Yes. The qdrant/qdrant repository was last updated on Oct 21, 2026, and the latest release, v1.19.2 on GitHub, came out on Oct 5, 2026.

How popular is Qdrant?

Qdrant has 35.4k GitHub stars as of Nov 1, 2026. It gained 129 new stars in the week of Oct 25–31, 2026, against a 12-week average of 136 a week. Its main package had about 4.5M downloads that week, +5% against its own 4-week average. That made it #19 among RAG, data and memory tools on the Devtoolwatch Board.

Qdrant vs pgvector: how do they compare?

Qdrant and pgvector are both tracked by Devtoolwatch as RAG, data and memory tools. Qdrant has 35.4k GitHub stars and pgvector has 23.5k. In the week of Oct 25–31, 2026, Qdrant's main package had 4.5M downloads and pgvector's had 9.4M. pgvector is an open-source Postgres extension for exact and approximate vector similarity search, storing vectors alongside the rest of your data. pgvector review.

Qdrant vs Milvus: how do they compare?

Qdrant and Milvus are both tracked by Devtoolwatch as RAG, data and memory tools. Qdrant has 35.4k GitHub stars and Milvus has 46.5k. In the week of Oct 25–31, 2026, Qdrant's main package had 4.5M downloads and Milvus's had 1.1M. Milvus review.

What are the best alternatives to Qdrant?

Other RAG, data and memory tools on the Devtoolwatch Board in the week of Oct 25–31, 2026, by momentum: XERJ (#1), Memvid (#2), LEANN (#3), Graphify (#4), MarkItDown (#5). The alternatives table above has their numbers.

Is Qdrant a database?

Yes. Qdrant is a vector similarity search engine and vector database. It stores points, which are vectors with an additional JSON payload, and provides an API to store, search, and manage them. Write-ahead logging keeps data persistent, even during power outages.

Can Qdrant run locally?

Yes. Qdrant runs locally with docker run -p 6333:6333 qdrant/qdrant, then you connect at http://localhost:6333. For in-process use without a server, Qdrant Edge stores and queries data locally and can sync with a Qdrant server.

How do I use Qdrant in Python?

Qdrant's Python client is the qdrant-client package, which needs Python 3.10 or later. Import QdrantClient from qdrant_client and create a client with QdrantClient(url="http://localhost:6333") to connect to a running server.

Can Qdrant be used for RAG?

Yes. Qdrant lists RAG and AI agents among its use cases. It provides vector search with metadata filtering, hybrid dense and sparse search, and reranking options such as late interaction models and MMR.

Does Qdrant support hybrid search?

Yes. Qdrant combines multiple vectors in one query and merges results with fusion strategies such as Reciprocal Rank Fusion and Distribution-Based Score Fusion. The homepage lists support for BM25, SPLADE++, and miniCOIL sparse representations.

What does Qdrant integrate with?

Qdrant integrates with embedding providers, AI application frameworks, data pipeline tools, observability platforms, and no-code and low-code automation platforms. The Ecosystem page in the documentation has the complete list.

Sources

Qdrant README badge

Maintainers can show the tool's weekly Board rank in their README:

[![Devtoolwatch](https://index.devtoolwatch.com/badge/qdrant.svg)](https://index.devtoolwatch.com/tools/qdrant/)

How this page is made: Devtoolwatch writes each review from the project's README, releases, package metadata and website, with an AI model, then checks every claim against the source it cites and drops what the sources don't support. The numbers come from Devtoolwatch's daily GitHub snapshots, GitHub's stargazer history, and the npm and PyPI download APIs. Spotted an error? Email [email protected].