Serverless Qdrant
Production-grade vector search with a real REST endpoint. Web search console, the same Qdrant client libraries, cloud or self-host.
/collections/products/points/searchBody
Response200 OK · 6ms
Pro engine
$15/mo flat, no meters
Encrypted
TLS on every plan
Layerbase Cloud
Managed, in seconds
Desktop and CLI
Run it locally or self-host
HTTPS endpoint
Qdrant over HTTPS, from anywhere
Qdrant on Layerbase Cloud is one HTTPS endpoint plus an API key from the dashboard. The official clients connect unchanged, and serverless or edge runtimes that cannot hold a TCP connection work the same way. The dashboard console is a thin layer over the same REST API, so what you see in the browser is what your code sees, and the firewall panel locks the endpoint to your own IPs or CIDR ranges. The same database branches copy-on-write, so an experiment or a CI job can have an isolated copy of your collections in seconds.
Serverless and edge accessQuick start
Connect with any Qdrant client
Vector search for RAG and semantic search, embedding-backed recommendations, multimodal retrieval, and apps that need real Qdrant features (payload filtering, hybrid search, recommendations).
- Stand up a Qdrant for a RAG prototype without managing a node
- Move from local Qdrant Docker to a hosted endpoint without changing client code
- Multi-tenant vector search where each tenant gets isolated collections
- HTTPS-only access from serverless or edge runtimes
- Self-host on your own hardware with the Layerbase CLI when latency requires it
curl -H "api-key: YOUR_TOKEN" "https://your-host.cloud.layerbase.dev/collections"
Where to run it
Cloud, desktop, or self-host
One database, three places it can live. Pick the speed you want and move between them when you outgrow it.
Managed Qdrant hosting
What managed Qdrant costs here, and how it runs
Qdrant is a Pro engine on Layerbase Cloud: the floor is the $15/mo Pro plan, flat, with no per-query and no per-vector meter. It runs always-on rather than sleeping, so a retrieval call never pays a wake penalty.
Qdrant needs the Pro plan, $15/mo
Pro is $15/mo flat: 10 databases, 25 GB of storage, up to 500 concurrent connections, and a 1.5 GB always-on pool. Creating a Qdrant draws 512 MB of that pool. Treat that as a floor rather than a budget, because a vector index grows resident with your data; pool blocks stack at $10/mo each for another 1 GB of RAM, 1 vCPU, and 25 GB of storage.
Always-on, so retrieval never waits on a cold start
Qdrant never hibernates here and is never auto-archived. That matters more for a vector store than for most engines: a RAG call that has to wake a database first is a user-visible pause in the middle of an answer.
HTTPS REST with the official clients
REST works out of the box and the official qdrant-client libraries for Python, Node, Go, and Rust connect to the published HTTPS endpoint with a dashboard-provided API key. gRPC is in active rollout; REST covers the same operations with comparable latency in most cases. A firewall panel restricts the endpoint to specific IPs or CIDR ranges, and the search console is a thin layer over the same REST API.
Branch a vector database
Qdrant is branchable here: take a copy-on-write copy of the running collection set in seconds, re-embed or re-index against the copy, compare results, and delete it when you are done. Pro allows 10 branches per database, and a branch does not consume a slot from your plan database count.
Nothing expires, and backups are automatic
Auto-archiving is a free-tier sweep, so a Pro database is never archived however long it sits. Pro takes a daily backup with 30 days of rolling retention, plus 10 manual slots you trigger yourself, and restore is one click from the dashboard onto the same database or into a new one.
Qdrant needs the Pro plan, $15/mo flat, with a 7-day free trial. Pool blocks add capacity at $10/mo.
FAQ
Common Qdrant questions
What is Qdrant?
Qdrant is an open-source vector database for similarity search over embeddings, with payload filtering, hybrid search, and recommendation APIs on top. It is a common retrieval layer behind RAG pipelines, semantic search, and embedding-backed recommendations. Layerbase runs it as a managed HTTPS REST endpoint with a web search console, and the official qdrant-client libraries connect with the dashboard-provided API key.
How much does managed Qdrant hosting cost?
Qdrant is a Pro engine on Layerbase Cloud, so the floor is the $15/mo Pro plan; it is not on the Free plan. Pricing is flat with no per-query or per-vector meter, and the cost preview in the create flow shows the monthly cost before you provision anything.
Does Layerbase Qdrant support REST or gRPC?
REST works out of the box on every plan. gRPC support is in active rollout (see status updates); REST covers the same operations with comparable latency in most cases.
Will my Qdrant Python client work?
Yes. The official qdrant-client (Python, Node, Go, Rust) connects to the published HTTPS endpoint with the dashboard-provided API key.
How do I manage collections?
Through the dashboard search console or via your client library directly. The console is a thin layer on the REST API and shows the same shape your code will see.
Can I restrict my Qdrant endpoint to specific IP addresses?
Yes. The firewall panel in the dashboard accepts IPs and CIDR ranges and locks the endpoint to those sources.
Is there a free Qdrant tier on Layerbase?
No. Qdrant is a Pro engine on Cloud, so the floor is the $15/mo Pro plan, and creating one draws 512 MB from your always-on pool. The Free plan covers the Standard engines for prototyping, and Qdrant is free to run locally with the Layerbase CLI or Layerbase Desktop.
Can I run Qdrant locally?
Yes. The Layerbase CLI and Layerbase Desktop both run Qdrant with the same one-command install and the same dashboard UI.
Branch Qdrant in seconds
Fork a running Qdrant into a full, isolated copy with its own connection string. Copy-on-write makes it near-instant, not a dump and restore. Point a preview deploy at it, test a migration, then throw it away. Layerbase branches 16 engines, not just Postgres.
See how branching worksTry Qdrant free for 7 days
Qdrant runs always-on from your Pro pool, so it lives on the Pro plan. Start a 7-day Pro trial to spin one up: a card is required, you are charged $0, and you can cancel before it converts to $15/mo. One trial per account.
Start the Pro trialWhat you get
Everything a Qdrant database should ship with
The same primitives on every Qdrant database: TLS, web console, hibernation, and a real dashboard. The rows that need a paid plan say so on the row; nothing here is hidden behind a sales call.
Web IDE
Query console in the dashboard, no extra client to install.
TLS by default
All connections encrypted. No setup required.
Always-on connections
Pin a database to your plan's always-on pool so it never hibernates. Free has no pool, so a free database always sleeps when idle; Solo includes 768 MB of pool and Pro 1.5 GB.
Scale to zero
Hibernates when idle. Wakes on the next connection.
Connection pooling
Pooled endpoint for serverless and edge workloads.
Direct connections
Bypass the pooler for migrations and replication.
Native HTTPS endpoint
Engines whose own wire protocol speaks HTTPS. Every database, including this one, is also queryable over the platform's HTTPS query API and web console.
Serverless driver
Drop-in HTTP driver for edge runtimes (Neon serverless, PlanetScale, Upstash REST, or the engine-native client).
IP whitelisting
Restrict access to specific IPs or CIDR ranges.
Automatic backups
Scheduled dumps and volume snapshots with one-click restore on paid plans. Free includes 1 manual backup slot you take and restore yourself.
What it costs
What Qdrant costs on Layerbase
Plans are flat. There is no per-hour compute meter, no per-database line item, and no ingress or egress fee, so the bill does not change shape when your traffic does. You do not need an AWS, GCP, or Azure account either: Layerbase runs the infrastructure.
Every engine, plus a bigger always-on pool.
Side by side
Hosted Qdrant, compared
| Provider | Free tier | Billing |
|---|---|---|
| Qdrant Cloud | Free cluster with 0.5 vCPU, 1 GB RAM and 4 GB disk | Paid clusters are billed hourly on the vCPU, memory and storage used |
| Layerbase | Qdrant is a Performance engine on Pro | $15/mo flat as 1 of 10 databases, always-on from your pool. No meters. |
Qdrant Cloud
- Free tier
- Free cluster with 0.5 vCPU, 1 GB RAM and 4 GB disk
- Billing
- Paid clusters are billed hourly on the vCPU, memory and storage used
Layerbase
- Free tier
- Qdrant is a Performance engine on Pro
- Billing
- $15/mo flat as 1 of 10 databases, always-on from your pool. No meters.
Competitor facts verified September 7, 2026 against each vendor's own pricing and documentation pages: Qdrant Cloud.
More on Qdrant
All articles- September 2, 2026
Pinecone vs Weaviate vs Qdrant in 2026: which vector database should you actually run?
A three-way comparison of the vector databases that keep showing up in RAG stacks: architecture, hosting model, current pricing, hybrid search, and multi-tenancy. Verified prices, no borrowed benchmark charts.
Read - August 15, 2026
Branching vector databases
Embeddings are expensive to compute and easy to ruin. Qdrant, Weaviate, and Meilisearch all branch on Layerbase now: fork the store per eval run, per agent, or per risky reindex, and throw the fork away.
Read - August 6, 2026
Qdrant 1.18 is on Layerbase, and the Headline is Quantization
Qdrant 1.18 is the new default line on Layerbase. Here is what a vector database is actually doing, why quantization decides what your index costs to run, and what landed between 1.16 and 1.18.
Read - June 20, 2026
Run Postgres, Valkey, and Qdrant in one Lovable app
A serious Lovable app usually needs more than one database: durable state, a cache, and vector search. Here is how to wire all three in, the correct way, from a browser SPA that has no server of its own.
Read - June 14, 2026
Branching with any database
Neon branches Postgres. PlanetScale branches MySQL. Layerbase branches all of them, because branching happens at the filesystem, not inside the engine. Here is how it works.
Read
Upstream project
Qdrant is developed by the Qdrant project. Layerbase packages and hosts official releases; the engine itself is built upstream.