Deploy and Host Qdrant on Railway

Qdrant is an open-source, high-performance vector database written in Rust for similarity search and retrieval-augmented generation (RAG). Store embeddings, filter on payloads, and run sub-millisecond nearest-neighbor search behind a REST + gRPC API — a self-hosted alternative to Pinecone and Weaviate.

About Hosting Qdrant

This template runs the official qdrant/qdrant image wired for Railway. Qdrant serves its REST API on port 6333 and gRPC on 6334; the template sets PORT=6333 so Railway's proxy targets the REST port and exposes it over HTTPS. Collections and the vector index live on an attached volume at /qdrant/storage, so your data survives redeploys. Every data route is protected by the QDRANT__SERVICE__API_KEY bearer token auto-generated at deploy — only /healthz, /livez, and /readyz stay open so Railway can healthcheck the service. A built-in web dashboard is available at /dashboard. Qdrant is memory-resident: RAM and disk scale with the number and dimensionality of your vectors, so size the service to your dataset.

Why Deploy Qdrant on Railway?

Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it.

By deploying Qdrant on Railway, you are one step closer to supporting a complete full-stack application with minimal burden. Host your servers, databases, AI agents, and more on Railway.

RAG memory in one click — give any LLM app long-term vector memory without a managed-service bill.
Your embeddings stay yours — vectors and payloads never leave your infrastructure; no per-vector SaaS fees.
Secure by default — API key auth is auto-generated; volume, healthcheck, and HTTPS domain are preconfigured.
Drop-in Pinecone alternative — point the official Python/JS/Rust/Go clients at your own domain.

Common Use Cases

Retrieval-augmented generation — store document embeddings and fetch relevant context for chatbots, agents, and copilots.
Semantic and hybrid search — power "search by meaning" over docs, products, or support tickets with payload filtering.
Recommendations and deduplication — nearest-neighbor lookups for similar items, images, or near-duplicate detection.

Dependencies for Qdrant Hosting

A volume mounted at /qdrant/storage (included in this template) for persistent collections and the vector index.
A service sized to your dataset — vectors are held in RAM for fast search.

Deployment Dependencies

Qdrant — upstream vector database by Qdrant (Apache-2.0)
Qdrant documentation — API reference and client SDKs
qdrant/qdrant Docker image — the official image this template deploys

Implementation Details

Create a collection, add vectors, and search:

# create a collection
curl -X PUT 'https:///collections/test' \
  -H 'api-key: $API_KEY' -H 'Content-Type: application/json' \
  -d '{"vectors":{"size":4,"distance":"Dot"}}'

# upsert a point
curl -X PUT 'https:///collections/test/points' \
  -H 'api-key: $API_KEY' -H 'Content-Type: application/json' \
  -d '{"points":[{"id":1,"vector":[0.1,0.2,0.3,0.4],"payload":{"t":"akira"}}]}'

# search
curl -X POST 'https:///collections/test/points/search' \
  -H 'api-key: $API_KEY' -H 'Content-Type: application/json' \
  -d '{"vector":[0.1,0.2,0.3,0.4],"limit":1}'

Web dashboard at /dashboard. Official clients for Python, JS/TS, Rust, Go, Java, and .NET.