Qdrant: semantic search on your own infrastructure
Production-grade vector database for semantic search
This template does not use a GPU: you can also run it on a dedicated CPU machine if you would rather have more cores and memory than a graphics card.
- Video memory
- não precisa de GPU
- Setup
- ~2 min
- Access
- port 6333
- Billing
- hourly, in BRL
Qdrant stores your base as vectors and returns the passage closest to a question in milliseconds. It is the storage layer of any serious RAG project.
Qdrant is the production-grade open-source vector database: stores millions of vectors, fast filtering, JSON payload, snapshots, replication. Foundation for any RAG pipeline, semantic search, recommendation. Web dashboard included.
What it is for
- Semantic search over a catalog, archive or support base
- Long-term memory for an AI agent
- Similarity recommendations with no external service
- Metadata filtering alongside the vector search
How to deploy
- Create your account and add balance (card or Pix, no subscription).
- In the console, pick the Qdrant template and a machine — the console hides the ones that do not meet the requirement.
- In about 2 minutes the setup finishes and the access address shows up in the panel, on port 6333.
Done? Just destroy the machine and billing stops with it. No contract, no minimum commitment.
FAQ
Does it need a GPU?
Not for the database itself — pick a CPU machine. The GPU comes in when generating the vectors.
How many vectors can it hold?
Millions, with the practical limit set by the machine's memory and disk.
Is there an API?
Yes, HTTP and gRPC, plus client libraries for the common languages.