QDRANT CLOUD

Build Fast, Accurate Vector Search at Scale with Qdrant Cloud

Qdrant Cloud is a managed service for running Qdrant vector search. You get the same open-source Rust engine with secure, predictable performance at scale. Move between self-hosted and managed without changing your code or data.

Start Free

Predictable P99 Latency

Native Hybrid + Multimodal

SOC 2 + HIPAA + GDPR

TRUSTED BY ENTERPRISE TEAMS

WHY QDRANT CLOUD

Offload Operations to Qdrant Cloud

Get the same predictably fast and accurate vector search engine. Qdrant Cloud handles the infrastructure.

Simplify Cluster Operations

Upgrades, scaling, sharding, backups, and monitoring run continuously without operator action.

Audited Compliance

Certified under SOC 2 Type II, HIPAA, and GDPR. BAA and DPA available on request.

No Vendor Lock-In

The same Qdrant engine, data format, and APIs across self-hosted, hybrid, and managed Cloud. Cloud supports snapshot export to your own infrastructure.

Start Free

IN PRODUCTION

Customers Building with Qdrant Cloud

2-3x Revenue Lift

Travelers using the generative AI experience show 2-3x more revenue than those using traditional search.

7x Faster Production Time

After replacing their vector stack with Qdrant, Deutsche Telekom cut AI agent development from 15 days to 2 days.

<1s Filtered query response on 14M vectors

Sub-second response times on complex, highly filtered semantic queries across an indexed archive of more than 14 million vectors.

Capabilities Included with Every Cloud Cluster

Each capability is available through the Qdrant Cloud Console and the API.

Composable Search for the Whole Retrieval Pipeline

You choose how each query is ranked, filtered, and scored. Combine dense vectors, sparse vectors, and metadata filters at query time.

Hybrid Search

Keyword and semantic search lives in the same engine. Dense and sparse vectors in one query. Native BM25 and SPLADE++ run alongside dense retrieval.

Filterable HNSW

Latency stays predictable under filters. Filtering integrates with graph traversal, beyond pre- and post-filtering tradeoffs.

Built-in Multivector

More precise multimodal search in one query. Store multiple vectors per object across text, image, audio, video.

Full-Spectrum Reranking

Apply business logic and token-level precision. Score boosting, ColBERT, and Maximum Marginal Relevance in-engine.

Advanced Metadata Filters

Enable more precise and efficient retrieval. Store metadata in JSON and use advanced filters, such as nested, text, geo, has_vector, and more.

Ingestion and Updates

Bulk upserts and streaming. Insert, update, and delete on a live index.

Qdrant Cloud Inference

Embed and query in one round trip. Native embedding generation inside a cluster.

Recommendation API

Get “more like this” with one API call. Positive and negative examples to find similar items.

Control Performance at Scale

The same engine runs from in-memory dev to web-scale production. Tune memory, indexing speed, and capacity for each workload.

Quantization

Up to 32× memory reduction. Scalar, TurboQuant, and binary help strike a balance between accuracy, storage efficiency, and search speed.

On-Disk Storage

Offload cold vectors and payloads to disk to reduce RAM cost.

Vertical and Horizontal Scaling

Shards rebalance automatically, maintaining optimal performance. Scale clusters up, down, or out.

GPU Indexing

Up to 4× faster HNSW indexing. Every node in your cluster gets a dedicated GPU with a simple toggle.

Flexible Multitenancy

Serve thousands of tenants per cluster with payload-based separation. Promote noisy ones to dedicated while traffic continues.

SIMD and Async I/O

SIMD acceleration across x86 and ARM. io_uring keeps disk throughput high on Cloud volumes.

High Availability and Recovery

Engine-handled operations on highly available clusters. Continuous backups and live upgrades.

Zero-Downtime Upgrades

Engine upgrades run while the cluster serves traffic. Multi-version supported.

Multi-AZ Replication

Up to 99.95% SLA. Three availability zones with automatic failover.

Backups and Disaster Recovery

Scheduled incremental backups, on-demand snapshots, and restore to any cluster.

Snapshot Export

Export snapshots to your own object storage for long-term retention or DR.

Engineering Support

Guaranteed response times on critical incidents. Business-hours coverage on Standard, 24/7 on Premium.

Defense in Depth for Production Workloads

Encryption on every channel and volume, granular access control, and private network options. Compliance-ready under SOC 2, HIPAA, and GDPR.

Compliance

Reports available under NDA. SOC 2 Type II, HIPAA with BAA, and GDPR with DPA.

Encryption in Transit

End-to-end protection in flight. TLS 1.2+ on every API endpoint and replication channel.

Encryption at Rest

Storage protected by default; Premium adds customer-managed keys. AES-256 on storage volumes.

Private VPC Links

Traffic stays on private networks. AWS PrivateLink and GCP Private Service Connect.

IP Allowlisting

Define your network perimeter explicitly. Restrict cluster access to specific CIDR ranges.

Single Sign-On (SSO)

Manage Cloud access through your existing identity provider. SAML 2.0 with Okta, Azure AD, Google and others.

Monitoring and Observability Toolkit

Monitor cluster health, audit every API call, and visualize capacity in real time. Configurable retention for compliance.

Prometheus Metrics

OpenMetrics-compatible /metrics and /sys_metrics endpoints on every cluster.

Grafana Dashboard

Reference dashboard for cluster health, query latency, and capacity.

Audit Logging

Logs every API operation: caller, target, and outcome in structured JSON.

Capacity Alerts

Alerts at 80% RAM and disk utilization, plus CPU throttling.

Health Endpoints

Kubernetes-style /healthz, /livez, /readyz on every node.

Telemetry

Per-segment, per-shard internals with configurable verbosity.

GETTING STARTED

Qdrant Cloud Across Your Stack

Language SDKs

Infrastructure as Code

Migrate

Integrations

Same workflow from your app, terminal, or CI.

Python SDK in app code, qcloud CLI in scripts and CI. Same Qdrant Cloud API.

Python SDK

from qdrant_client import QdrantClient

client = QdrantClient(
    url="https://your-cluster.qdrant.io",
    api_key="qdrant_…",
)

# List collections on this cluster
catalog = client.get_collections()

# Snapshot before risky index or schema changes
client.create_snapshot(collection_name="products")

# Page payloads for spot checks, exports, or pipeline validation
points, next_offset = client.scroll(
    collection_name="reports",
    limit=100,
    with_payload=True,
)

qcloud CLI · REST

# Pick the Qdrant Cloud context (API key + account id)
qcloud context use prod

# Inspect clusters — ids, regions, endpoints, status
qcloud cluster list
qcloud cluster describe $QDRANT_CLUSTER_ID

# Snapshot the collection (in-cluster) before risky index / schema changes
qcloud cluster snapshot create products

# Query data via REST when scripting outside the SDK
curl -sS "https://abcd-1234.eu-central.aws.cloud.qdrant.io:6333/collections/products/points/scroll" \
  -H "api-key: $QDRANT_DATA_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{"limit": 100, "with_payload": true, "with_vector": false}'

See what's in your cluster from the browser.

Inspect collections, run queries, and check results without leaving the Console.

Define Qdrant Cloud clusters in your IaC repo.

Provision, scale, and tear down clusters alongside the rest of your infrastructure.

terraform {
  required_providers {
    qdrant-cloud = {
      source  = "qdrant/qdrant-cloud"
      version = "~> 1.0"
    }
  }
}

provider "qdrant-cloud" {
  api_key    = var.qdrant_api_key
  account_id = var.qdrant_account_id
}

resource "qdrant-cloud_cluster" "production" {
  name              = "search-production"
  cloud_provider    = "aws"
  cloud_region      = "eu-central-1"

  configuration {
    number_of_nodes = 3
    node_configuration {
      package_id = "gpxxl-1"  # 8 vCPU, 32 GB RAM per node
    }
  }
}

# Pin the data API key to a Terraform output for downstream consumers
output "cluster_endpoint" {
  value = qdrant-cloud_cluster.production.url
}

TRY QDRANT CLOUD

Spin Up a Qdrant Cluster

Try Cloud Free

Spin up a permanently free cluster in under 90 seconds. No credit card, no minimum spend, around one million 768-dimension vectors.

Start Free

Talk to a Solutions Engineer

For SOC 2 evidence, HIPAA BAA, sizing help, or enterprise procurement.

Talk to Engineering

Run Open Source Locally

Pull the Docker image or clone the repo. Same engine, same APIs as Cloud. Apache 2.0.

View on GitHub

FAQs

Is there a free tier?

Yes. The free tier is a single-node cluster with 0.5 vCPU, 1GB RAM, and 4GB disk. That fits roughly one million 768-dimension vectors. No credit card. Free clusters suspend after 1 week of inactivity and delete after 4 weeks if you don't reactivate them.

What clouds and regions are supported?

AWS, GCP, and Azure across multiple regions. The current questions shows up in the cluster creation flow and on the pricing page. New regions get added when customers ask for them.

Which clients and SDKs do you support?

Official SDKs for Python, TypeScript, Rust, Go, Java, and .NET. The REST and gRPC APIs are documented and identical to open-source Qdrant, so any community client built against the OSS engine works against Cloud.

Can I bring my own embedding model?

Yes. Qdrant Cloud is model-agnostic, so you can bring vectors from OpenAI, Cohere, Voyage, or your own fine-tunes. If you want one less vendor, Qdrant Cloud Inference generates text and image embeddings inside the cluster using models like MiniLM, SPLADE, BM25, Mixedbread Embed-Large, and CLIP. Paid clusters get up to 5 million free tokens per model per month, with no token cap on BM25.

How is pricing calculated?

Resource-based: vCPU, RAM, and storage, billed hourly. Cloud Inference adds usage charges only when you call paid embedding models. The pricing page has a sizing calculator.

Are there overage charges or surprise bills?

No. Cluster sizes are explicit and any change needs your authorization. Capacity alerts fire at 80% so you can upgrade on your own terms before anything breaks.

Do you offer startup or research discounts?

Yes. The Qdrant for Startups program offers a 20% Qdrant Cloud discount for 12 months. Eligibility: pre-seed, seed, or Series A, under 5 years old, under $5M in funding, building an AI product (agencies and dev shops don't qualify). Apply at qdrant.tech/qdrant-for-startups.

What query latency should I expect?

For a typical 10M-vector collection at 768 dimensions with moderate filtering, P50 lands in the single-digit milliseconds and P99 stays under 50ms. Numbers shift with dimension count, recall target, filter selectivity, and cluster size. Reproducible benchmarks at qdrant.tech/benchmarks.

How fast can I ingest data?

CPU indexing handles tens of thousands of vectors per second per node, depending on dimension and HNSW parameters. GPU-accelerated HNSW indexing delivers up to 4x faster index construction for bulk loads. GPU clusters are available today on AWS, with other clouds on the roadmap.

How does quantization affect recall?

Scalar quantization typically loses 1% to 2% recall and cuts RAM by 4x. TurboQuant (Google's algorithm, integrated into Qdrant) sits in the middle: 4-bit gives roughly 8x compression with recall close to scalar, up to 32x at higher ratios. Binary with rescoring keeps recall close to full precision while cutting memory by up to 32x for compatible embedding models. The cluster UI shows recall before and after so you can pick the trade-off you want.

How do I scale up or down?

Vertical scaling adds vCPU, RAM, or disk to existing nodes. Horizontal scaling adds nodes and rebalances shards automatically. Both run from the dashboard or the API. If your collections aren't replicated, a vertical scale takes a short downtime window. Replicated collections scale without interruption.

What happens when my cluster gets full?

If RAM or disk usage stays above 80% for 5 minutes, the account owner gets an email alert. From there you can scale vertically (more capacity per node) or horizontally (more nodes), or delete data. No surprise lockouts, no surprise bills.

Is multi-region supported?

Multi-AZ within a single region is available on the Premium Multi-AZ tier. It needs a minimum of three nodes and scales in multiples of three. Multi-region active-active replication isn't a managed feature yet. Teams that need it run separate clusters per region today.

What's your SLA?

99.5% uptime on Standard. 99.9% on Premium. Up to 99.95% on Premium Multi-AZ. The full SLA, including uptime definitions and service-credit terms, lives at qdrant.to/sla.

How do backups work?

Scheduled incremental snapshots on AWS and GCP, with configurable retention. Azure backups bill based on total disk usage. You can also take on-demand snapshots before risky changes and restore to the same cluster or a new one. For long-term retention or compliance, you can export snapshots to your own object storage.

What's your disaster recovery posture?

You set the snapshot cadence to match your RPO. Premium Multi-AZ deployments replicate across three availability zones (cross-AZ replication, not failover) with no failover delay. If a zone goes down, reads and writes continue from the surviving zones with no customer action required. For cross-region retention, export snapshots to your own object storage in any region.

Are you SOC 2, GDPR, and HIPAA compliant?

Yes. SOC 2 Type II report and HIPAA certification on file. GDPR-compliant Data Processing Agreement available. Email Solutions Engineering for current compliance documentation and BAA scope.

How is data encrypted?

TLS in transit. Storage volumes encrypted at rest. Premium customers can use customer-managed keys for disk encryption, plus SSO and VPC private links. Snapshots and backups inherit the same encryption.

What does audit logging capture?

Every API operation: queries, upserts, deletes, collection management, and snapshot operations. Each entry is structured JSON with caller identity, timestamp, target collection, and the decision (allowed or denied). You can retrieve logs through an API endpoint, configure retention to match your policy, and download them for long-term storage in your own systems. Available on all paid clusters.

Are you EU-based?

Yes. Qdrant is headquartered in Berlin. EU customers can keep data in EU regions exclusively, which addresses US Cloud Act concerns and similar extraterritorial regimes.

Is migrating from Qdrant Open Source to Cloud difficult?

No. Our open-source migration tool turns it into a configuration change, not a rewrite. Application code points at a new endpoint; business logic stays intact.

How do I migrate from another vector database?

The migration tool covers Pinecone, Weaviate, Milvus, Chroma, Redis, MongoDB, OpenSearch, Elasticsearch, pgvector, S3 Vectors, FAISS, Apache Solr, and Qdrant-to-Qdrant (for example, OSS to Cloud). The bigger lift is usually re-running the ingestion pipeline; the data move itself is incremental and resumes if interrupted. Solutions Engineering will pair on a migration plan if you ask.

Can I move workloads back from Cloud to OSS later?

Yes. Export a snapshot, restore it on your own infrastructure running open-source Qdrant. The engine and data format are identical. Your data is yours.

What happens if Qdrant Cloud goes away?

The engine is open source under Apache 2.0, with 30k+ GitHub stars and a 60k-member Discord community. You can run it yourself indefinitely. Engine parity keeps your options open whatever happens to the managed service.

What support tiers do you offer?

Community (Discord, free), Standard (10x5 business hours, Mon-Fri 08:00 to 18:00 CET), and Premium (24x7 critical incident response with priority response times).

How fast do you respond?

We use four severity levels. Standard customers get a Sev 1 response in 4 business hours, Sev 2 in 6, Sev 3 in 24. Premium customers get Sev 1 in 1 hour, Sev 2 in 2, Sev 3 in 4 business hours. Real engineers, not a tier-1 chatbot.

Do you have a community channel?

Yes. The Qdrant Discord is open to all developers, including the engineering team. discord.gg/qdrant

Build Fast, Accurate Vector Search at Scale with Qdrant Cloud

Offload Operations to Qdrant Cloud

Simplify Cluster Operations

Audited Compliance

No Vendor Lock-In

Customers Building with Qdrant Cloud

2-3x Revenue Lift

7x Faster Production Time

<1s Filtered query response on 14M vectors

Capabilities Included with Every Cloud Cluster

Composable Search for the Whole Retrieval Pipeline

Hybrid Search

Filterable HNSW

Built-in Multivector

Full-Spectrum Reranking

Advanced Metadata Filters

Ingestion and Updates

Qdrant Cloud Inference

Recommendation API

Control Performance at Scale

Quantization

On-Disk Storage

Vertical and Horizontal Scaling

GPU Indexing

Flexible Multitenancy

SIMD and Async I/O

High Availability and Recovery

Zero-Downtime Upgrades

Multi-AZ Replication

Backups and Disaster Recovery

Snapshot Export

Engineering Support

Defense in Depth for Production Workloads

Compliance

Encryption in Transit

Encryption at Rest

Private VPC Links

IP Allowlisting

Single Sign-On (SSO)

Monitoring and Observability Toolkit

Prometheus Metrics

Grafana Dashboard

Audit Logging

Capacity Alerts

Health Endpoints

Telemetry

Qdrant Cloud Across Your Stack

Same workflow from your app, terminal, or CI.

See what's in your cluster from the browser.

Define Qdrant Cloud clusters in your IaC repo.

Pricing Based on Cluster Resources

Spin Up a Qdrant Cluster

Try Cloud Free

Talk to a Solutions Engineer

Run Open Source Locally

FAQs

Is there a free tier?

What clouds and regions are supported?

Which clients and SDKs do you support?

Can I bring my own embedding model?

How is pricing calculated?

Are there overage charges or surprise bills?

Do you offer startup or research discounts?

What query latency should I expect?

How fast can I ingest data?

How does quantization affect recall?

How do I scale up or down?

What happens when my cluster gets full?

Is multi-region supported?

What's your SLA?

How do backups work?

What's your disaster recovery posture?

Are you SOC 2, GDPR, and HIPAA compliant?

How is data encrypted?

What does audit logging capture?

Are you EU-based?

Is migrating from Qdrant Open Source to Cloud difficult?

How do I migrate from another vector database?

Can I move workloads back from Cloud to OSS later?

What happens if Qdrant Cloud goes away?

Considering a
Production Deployment?