The Qdrant Vector Database Workbench for VS Code

Manage collections · Inspect points · Run vector search · Debug REST payloads — without leaving your editor.

Overview

Qdrant DB Workbench is a VS Code-native GUI for Qdrant vector databases. Connect to local Docker, Qdrant Cloud, or any Qdrant-compatible REST endpoint and inspect collections, points, vectors, payloads, and search responses from a focused editor panel.

It is built for developers testing embeddings, RAG pipelines, semantic search, and production vector search APIs.

Features

	Capability	Description
🔌	Universal Connect	Connect to local Qdrant, Qdrant Cloud, or a custom REST endpoint
🗂️	Collection Explorer	Browse collections with vector config, distance metric, and point counts
📍	Point Inspector	Retrieve and inspect point IDs, payloads, and vectors
🔍	Similarity Search	Run vector search with top K, filters, payload, and vector options
🧰	Raw REST Console	Send custom requests to any Qdrant REST path
🛡️	API Key Ready	Use optional API keys for secured cloud or private deployments
🌉	Extension-Host Bridge	Local and cloud HTTP calls flow through the extension host to avoid webview CORS issues
⚡	No Config Files	Enter the endpoint and connect directly from the workbench

Quick Start

Step 1 — Install

Search Qdrant DB Workbench in the Extensions panel (Ctrl+Shift+X) or run:

ext install harishkaparwan.qdrant-db

Step 2 — Start Qdrant locally (skip if already running or using Cloud)

docker run --rm -p 6333:6333 -p 6334:6334 qdrant/qdrant:latest

Check the service:

curl http://localhost:6333/

Step 3 — Open the Workbench

Click the Qdrant DB icon in the Activity Bar, or open the Command Palette and run:

Qdrant DB Workbench: Open Workbench

Step 4 — Connect

Mode	What to enter
Local	`http://localhost:6333`
Cloud	Qdrant Cloud URL plus API key

Connection Modes

Local

Use a local Qdrant REST endpoint:

http://localhost:6333

Works with Docker, local binaries, or a private proxy.

Cloud

Use your Qdrant Cloud cluster URL and API key:

https://<cluster-id>.<region>.cloud.qdrant.io

Credentials are entered in the workbench UI and should not be written into source files.

Workbench Tabs

Tab	Use it for
Overview	Service status, selected collection, and high-level metrics
Collections	Browse collections, vector configuration, and collection details
Points	Fetch and inspect points, payloads, and vectors
Query	Run vector similarity searches with filter payloads
Raw	Send direct REST requests to advanced Qdrant endpoints

Works With VS Code-Compatible Editors

Editor	Platform	Notes
VS Code	Desktop · Web	Primary supported editor
Cursor	Desktop	VS Code-compatible AI editor
Windsurf	Desktop	VS Code-compatible AI editor
VSCodium	Desktop	Open-source VS Code build
Gitpod	Cloud	Browser-based development environments
GitHub Codespaces	Cloud	GitHub cloud IDE
Eclipse Theia	Cloud · Desktop	Open-source VS Code API-compatible IDE