Vector Databases API Key

Qdrant REST API

High-performance vector similarity search engine

Qdrant is an open-source vector database and similarity search engine optimized for neural network embeddings. It provides fast and scalable vector search with advanced filtering capabilities, making it ideal for semantic search, recommendation systems, and AI applications. Developers use Qdrant to store, search, and manage high-dimensional vectors with production-ready performance and flexible deployment options.

Base URL https://api.qdrant.tech:6333

API Endpoints

Method	Endpoint	Description
GET	`/collections`	List all collections in the database with their configurations and statistics
POST	`/collections/{collection_name}`	Create a new collection with specified vector size and distance metric
GET	`/collections/{collection_name}`	Get detailed information about a specific collection including size and configuration
DELETE	`/collections/{collection_name}`	Delete an existing collection and all its points
PUT	`/collections/{collection_name}/points`	Insert or update points (vectors) in a collection with optional payload data
POST	`/collections/{collection_name}/points/search`	Search for similar vectors in a collection with optional filters and scoring
POST	`/collections/{collection_name}/points/scroll`	Retrieve points from a collection with pagination and filtering
GET	`/collections/{collection_name}/points/{id}`	Retrieve a specific point by its ID including vector and payload
POST	`/collections/{collection_name}/points/delete`	Delete points from a collection by IDs or filter conditions
POST	`/collections/{collection_name}/points/recommend`	Get recommendations based on positive and negative example vectors
PUT	`/collections/{collection_name}/index`	Create or update an index for payload fields to enable fast filtering
POST	`/collections/{collection_name}/snapshots`	Create a snapshot of the collection for backup or transfer purposes
GET	`/collections/{collection_name}/snapshots`	List all available snapshots for a collection
GET	`/cluster`	Get cluster status and information about distributed deployment
GET	`/health`	Check the health status of the Qdrant instance

Code Examples

# Create a collection
curl -X POST 'https://api.qdrant.tech:6333/collections/my_collection' \
  -H 'api-key: your_api_key_here' \
  -H 'Content-Type: application/json' \
  -d '{
    "vectors": {
      "size": 384,
      "distance": "Cosine"
    }
  }'

# Insert points with vectors
curl -X PUT 'https://api.qdrant.tech:6333/collections/my_collection/points' \
  -H 'api-key: your_api_key_here' \
  -H 'Content-Type: application/json' \
  -d '{
    "points": [
      {
        "id": 1,
        "vector": [0.05, 0.61, 0.76, 0.74],
        "payload": {"city": "Berlin", "country": "Germany"}
      }
    ]
  }'

# Search for similar vectors
curl -X POST 'https://api.qdrant.tech:6333/collections/my_collection/points/search' \
  -H 'api-key: your_api_key_here' \
  -H 'Content-Type: application/json' \
  -d '{
    "vector": [0.2, 0.1, 0.9, 0.7],
    "limit": 5,
    "with_payload": true
  }'

// Create a collection
const createCollection = async () => {
  const response = await fetch('https://api.qdrant.tech:6333/collections/my_collection', {
    method: 'POST',
    headers: {
      'api-key': 'your_api_key_here',
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      vectors: {
        size: 384,
        distance: 'Cosine'
      }
    })
  });
  return await response.json();
};

// Insert points
const insertPoints = async () => {
  const response = await fetch('https://api.qdrant.tech:6333/collections/my_collection/points', {
    method: 'PUT',
    headers: {
      'api-key': 'your_api_key_here',
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      points: [
        {
          id: 1,
          vector: [0.05, 0.61, 0.76, 0.74],
          payload: { city: 'Berlin', country: 'Germany' }
        }
      ]
    })
  });
  return await response.json();
};

// Search vectors
const searchVectors = async () => {
  const response = await fetch('https://api.qdrant.tech:6333/collections/my_collection/points/search', {
    method: 'POST',
    headers: {
      'api-key': 'your_api_key_here',
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      vector: [0.2, 0.1, 0.9, 0.7],
      limit: 5,
      with_payload: true
    })
  });
  return await response.json();
};

import requests
import json

base_url = 'https://api.qdrant.tech:6333'
headers = {
    'api-key': 'your_api_key_here',
    'Content-Type': 'application/json'
}

# Create a collection
def create_collection():
    response = requests.post(
        f'{base_url}/collections/my_collection',
        headers=headers,
        json={
            'vectors': {
                'size': 384,
                'distance': 'Cosine'
            }
        }
    )
    return response.json()

# Insert points
def insert_points():
    response = requests.put(
        f'{base_url}/collections/my_collection/points',
        headers=headers,
        json={
            'points': [
                {
                    'id': 1,
                    'vector': [0.05, 0.61, 0.76, 0.74],
                    'payload': {'city': 'Berlin', 'country': 'Germany'}
                }
            ]
        }
    )
    return response.json()

# Search vectors
def search_vectors():
    response = requests.post(
        f'{base_url}/collections/my_collection/points/search',
        headers=headers,
        json={
            'vector': [0.2, 0.1, 0.9, 0.7],
            'limit': 5,
            'with_payload': True
        }
    )
    return response.json()

# Get collection info
def get_collection_info():
    response = requests.get(
        f'{base_url}/collections/my_collection',
        headers=headers
    )
    return response.json()

Use Qdrant from Claude / Cursor / ChatGPT

Get a hosted MCP endpoint for Qdrant. Paste your Qdrant API key, copy back one URL, drop it into Claude Desktop, Cursor, or any AI client that supports remote MCP. Your AI calls Qdrant directly with your credentials — no local install, works on mobile.

search_similar_vectors Search for semantically similar vectors in a collection using embedding-based similarity with optional metadata filtering

store_embeddings Store document or text embeddings with associated metadata in a specified collection for later retrieval

manage_collections Create, configure, and manage vector collections with specific distance metrics and indexing strategies

recommend_similar_items Get AI-powered recommendations based on positive and negative examples using vector similarity

filter_and_search Perform hybrid search combining vector similarity with structured payload filtering for precise results

Connect in 60 seconds

Paste your Qdrant key → get an MCP URL → paste into Claude/Cursor. Hosted by IOX, encrypted at rest.

Connect Qdrant to your AI →

Qdrant REST API

API Endpoints

Sponsor this page

Code Examples

Use Qdrant from Claude / Cursor / ChatGPT

Connect in 60 seconds

Related APIs