qdrant-mcp-server
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@qdrant-mcp-serverfind similar records about refund policy in the support collection"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
qdrant-mcp-server
A generic MCP server for Qdrant — store, search, list, and delete over any collection.
Unlike qdrant/mcp-server-qdrant, this server uses a plain (unnamed) vector, so it works out of the box with collections created the "default" way — e.g. by LangChain's or n8n's Qdrant node — instead of requiring the named-vector convention (fast-{model}) the official server hardcodes.
Features
Multi-collection:
QDRANT_COLLECTIONis only a default. Every tool accepts an optionalcollection_name, so one running instance can serve multiple collections.Plain vectors: compatible with collections populated by other tools/pipelines, not just this server.
OpenAI embeddings, with automatic token-aware truncation of oversized input.
Related MCP server: Qdrant MCP Server
Tools
Tool | Description |
| Embed and store a piece of text, with optional metadata and a stable |
| Semantic search over stored content. |
| Delete a point by the |
| List all collections available on the Qdrant server. |
Configuration
Set these environment variables (see .env.example):
Variable | Required | Description |
| yes | Qdrant server URL. |
| no | Qdrant API key, if auth is enabled. |
| no | Default collection name. Omit to require |
| yes | Used to generate embeddings. |
| no | Default |
| no | Default |
Running
Docker
docker build -t qdrant-mcp-server .
docker run --rm -p 8000:8000 --env-file .env qdrant-mcp-serverLocally
pip install -r requirements.txt
export $(cat .env | xargs) # or set the variables another way
python server.pyThe server speaks MCP over SSE on 0.0.0.0:8000.
Using with an MCP client
Point your client at the SSE endpoint, e.g. for a docker-compose service named mcp-qdrant:
{
"mcpServers": {
"qdrant": {
"url": "http://localhost:8000/sse"
}
}
}License
This server cannot be deployed
Maintenance
Related MCP Connectors
Ingest, manage, and retrieve documents for RAG-powered AI applications
Universal persistent memory and knowledge retrieval layer for AI agents and LLMs.
Semantic nearest-neighbour search over your datasets, plus indices and projections.
Persistent memory for AI agents. Search and store durable facts, preferences and decisions.
Related MCP Servers
- AlicenseNot gradedqualityNot gradedmaintenanceProvides RAG capabilities for semantic document search using Qdrant vector database and Ollama/OpenAI embeddings, allowing users to add, search, list, and delete documentation with metadata support.16 npm16-
- AlicenseNot gradedqualityCmaintenanceEnables semantic search and document management using a local Qdrant vector database with OpenAI embeddings. Supports natural language queries, metadata filtering, and collection management for AI-powered document retrieval.76 npm37MIT
- AlicenseNot gradedqualityDmaintenanceEnables storing and retrieving information using semantic search with Qdrant vector database. Acts as a memory layer for LLMs to persistently store and semantically search through information and metadata.Apache 2.0
- FlicenseNot gradedqualityDmaintenanceEnables interaction with Qdrant vector database for storing, searching, and managing vectors with automatic text embedding.1-