After you provision Qdrant hosting on RoamerHost, the project detail page in the dashboard shows two things: the REST API endpoint (in the form https://qdr-xxxx.roamerhost.com) and your API key. Every request must carry the api-key header; requests without a key are always rejected.
1. Verify the connection
curl https://qdr-xxxx.roamerhost.com \
-H "api-key: YOUR_API_KEY"If you see {"title":"qdrant - vector search engine", ...}, you're connected.
2. Create a collection
A collection is the equivalent of a table. size must match the output dimension of the embedding model you use (OpenAI text-embedding-3-small is 1536, multilingual E5 is 768):
curl -X PUT https://qdr-xxxx.roamerhost.com/collections/docs \
-H "api-key: YOUR_API_KEY" -H "Content-Type: application/json" \
-d '{"vectors": {"size": 1536, "distance": "Cosine"}}'3. Write vectors
In practice the vectors come from an embedding API; this shows the write format. payload holds the source text or any extra fields, and it comes back with the search hit:
curl -X PUT "https://qdr-xxxx.roamerhost.com/collections/docs/points?wait=true" \
-H "api-key: YOUR_API_KEY" -H "Content-Type: application/json" \
-d '{"points": [{"id": 1, "vector": [0.05, 0.61, ...], "payload": {"text": "Return policy: within seven days of receiving the item..."}}]}'4. Search
curl -X POST https://qdr-xxxx.roamerhost.com/collections/docs/points/search \
-H "api-key: YOUR_API_KEY" -H "Content-Type: application/json" \
-d '{"vector": [0.04, 0.59, ...], "limit": 3, "with_payload": true}'Results come back sorted by similarity; a higher score means more relevant.
Using Python
pip install qdrant-clientfrom qdrant_client import QdrantClient
client = QdrantClient(
url="https://qdr-xxxx.roamerhost.com",
api_key="YOUR_API_KEY",
)
print(client.get_collections())Everything after that — upsert, search — is exactly as described in the official documentation. RoamerHost runs the standard open-source Qdrant, with no API differences.
Related articles
- Building a RAG knowledge base with n8n + Qdrant: an AI Q&A bot with no code
- Qdrant explained: what is a vector database, and why does every AI application need one?
Want someone to build it for you?
If you'd rather not assemble these workflows yourself, or the scale is big enough that you want someone planning alongside you, Roamer Tech (RoamerHost's parent company) takes on contract work for business process automation and AI agent development:
