Providers
Hetzner
German infrastructure provider with low-cost cloud servers, load balancers, volumes and S3-compatible object storage; no managed Kubernetes and no H100/H200/L40S GPUs.
Last verified: Editorial review www.hetzner.com
- Full regions
- 3
- Cloud locations in Germany (Falkenstein, Nuremberg), Finland (Helsinki), the United States
- GA services
- 8
- of 8 listed
- Billing model
- Per hour
- EUR · USD
- H100 / H200 / L40S
- 0 / 3
- GPU models generally available
Overview
Hetzner Online offers Hetzner Cloud (cloud servers, load balancers, volumes, private networks, firewalls, primary and floating IPs, snapshots, backups and S3-compatible object storage) alongside dedicated root servers. Hetzner does not operate a managed Kubernetes control plane; it maintains a cloud-controller-manager and CSI driver for self-managed clusters. GPU capacity is offered as dedicated GEX servers with NVIDIA RTX PRO Blackwell GPUs rather than H100, H200 or L40S. Cloud servers are billed per started hour up to a monthly cap.
Compiled from Hetzner documentation and product pages listed in Sources. Audit-logging capabilities were not assessed and are recorded as unverified. Third-party reviews reported 2026 price increases and order restrictions for some new cloud customers; confirm current terms directly with Hetzner.
- Headquarters
- Gunzenhausen, Germany
- Billing model
- Cloud servers are billed per started hour up to a monthly price cap. Object Storage has an hourly base price with included storage and egress quotas.
- Technologies
- hcloud APITerraform (hcloud provider)Kubernetes CCM/CSIS3 API
Infrastructure regions
Cloud locations in Germany (Falkenstein, Nuremberg), Finland (Helsinki), the United States and Singapore; none in Türkiye.
| Region | Location | Type | Status | Notes |
|---|---|---|---|---|
| Falkenstein fsn1 | Falkenstein, Germany | Region | Generally available | |
| Nuremberg nbg1 | Nuremberg, Germany | Region | Generally available | |
| Helsinki hel1 | Helsinki, Finland | Region | Generally available |
Products and services
Compute
Cloud Servers
Generally availableShared and dedicated vCPU cloud servers (x86 and Arm).
Dedicated resources
Dedicated root servers
Generally availableBare-metal dedicated servers.
GPU & AI compute
GEX GPU dedicated servers
Generally availableDedicated servers with NVIDIA RTX PRO 4000 / 6000 Blackwell GPUs and setup fees.
Block storage
Volumes
Generally availableAttachable block storage volumes.
Object storage
Object Storage
Generally availableS3-compatible object storage in EU locations.
Backup
Backups and Snapshots
Generally availableServer backups and server snapshots.
Networking
Load Balancers
Generally availableLoad balancers for TCP, HTTP and HTTPS.
Networks, Firewalls and Floating IPs
Generally availablePrivate networks, stateful firewalls, primary and floating IPs.
GPU availability
| GPU | Status | Offering | List price |
|---|---|---|---|
| NVIDIA H100 | Not offered | Not offered | Not published |
| NVIDIA H200 | Not offered | Not offered | Not published |
| NVIDIA L40S | Not offered | Not offered | Not published |
Dedicated GEX servers with NVIDIA RTX PRO 4000 Blackwell (24 GB) and RTX PRO 6000 Blackwell Max-Q (96 GB) GPUs.
Capability matrix
The same criteria are applied to every provider. Values show documented capabilities, not performance scores.
| Criterion | Hetzner |
|---|---|
| Data residency | |
| Infrastructure located in Türkiye Generally available compute and storage capacity physically located in Türkiye (full region or local zone). | No |
| Compute & GPU | |
| Virtual machine instances Self-service virtual machines with selectable CPU/RAM profiles. | Yes |
| Per-second or per-minute billing Standard compute billed at per-second or per-minute granularity rather than per started hour. | No Per started hour with a monthly cap. |
| Dedicated / single-tenant resources Dedicated hosts, bare metal or organisation-dedicated cloud capacity. | Yes Dedicated root servers. |
| NVIDIA H100 instances On-demand or reservable instances with NVIDIA H100 GPUs. | No |
| NVIDIA H200 instances On-demand or reservable instances with NVIDIA H200 GPUs. | No |
| NVIDIA L40S instances On-demand or reservable instances with NVIDIA L40S GPUs. | No |
| Platform & automation | |
| API and Terraform automation Documented API plus an available Terraform provider for infrastructure as code. | Yes |
| Open-source cloud stack (OpenStack-based) The provider publicly states that its IaaS control plane is built on OpenStack or an equivalent open-source stack. | No |
| Kubernetes | |
| Managed Kubernetes Provider-operated Kubernetes control plane. | No CCM and CSI driver for self-managed clusters only. |
| Kubernetes node autoscaling Worker node count scales automatically within configured limits. | Partial Through the upstream cluster autoscaler on self-managed clusters. |
| Kubernetes LoadBalancer integration Services of type LoadBalancer provision a provider load balancer automatically. | Partial Via the Hetzner cloud-controller-manager on self-managed clusters. |
| Kubernetes storage classes (CSI) Persistent volumes provisioned dynamically through a provider storage class. | Partial Via the Hetzner CSI driver on self-managed clusters. |
| Storage & data | |
| Block storage volumes Persistent, attachable and expandable block volumes. | Yes |
| Volume snapshots Point-in-time snapshots of block volumes. | No Server snapshots only; no volume snapshots. |
| S3-compatible object storage Object storage reachable through the S3 API or an S3-compatible API. | Yes EU locations; partial S3 API coverage. |
| Managed backup service Scheduled backup and restore for instances or volumes managed by the provider. | Yes Server backups. |
| Managed database Provider-operated relational database service. | No |
| Networking | |
| L4 load balancer (TCP/UDP) Transport-layer load balancing. | Partial TCP only (no UDP). |
| L7 load balancer (HTTP/HTTPS) Application-layer load balancing with HTTP health checks. | Yes |
| Reassignable static public IP Static public addresses that can be moved between resources (Floating/Reserved/Elastic IP). | Yes |
| Private networks / VPC Isolated tenant networks with custom CIDR ranges. | Yes |
| Virtual router with NAT Managed router or NAT gateway for egress and inter-subnet routing, including static routes. | No Network routes are supported; NAT requires a self-managed gateway server. |
| Security groups / cloud firewall Stateful, API-managed network access rules. | Yes Stateful firewalls. |
| Security & auditing | |
| Control-plane audit logging A provider service that records management (API/console) operations against cloud resources. | Unverified |
| User identity attribution in audit events Each audit event identifies the authenticated principal that performed the operation. | Unverified |
Published benchmarks
Comparisons
Frequently asked questions
Does Hetzner offer managed Kubernetes?
No. Hetzner provides a cloud-controller-manager and CSI driver for clusters you run yourself; the control plane is not managed by Hetzner.
Sources
- [1] Hetzner Cloud — Hetzner Official documentation accessed:
- [2] Object Storage overview (Hetzner Docs) — Hetzner Official documentation accessed:
- [3] Dedicated servers with GPU (GEX) — Hetzner Official documentation accessed:
- [4] Hetzner Cloud Review 2026 — C9 Group Secondary source accessed:
Research references
Provider discovery
AWS alternatives with infrastructure in Türkiye
AWS, PortvMind Cloud, Google Cloud, Azure and the other tracked providers when a workload needs an AWS alternative in Türkiye.
Cloud security & auditing
AWS CloudTrail alternatives: audit logging on other clouds and OpenStack
CloudTrail's equivalents on Google Cloud, Azure, OpenStack-based clouds and smaller providers, compared on scope, identity, retention, export and integrity.
Provider discovery
Cloud providers in Türkiye: regions, local zones and GPU availability
Which global and regional cloud providers operate infrastructure in Türkiye, what type it is, and which services and GPUs are available locally.
Pricing
GPU cloud cost comparison: H100, H200 and L40S prices per GPU-hour
Published H100, H200 and L40S cloud prices per GPU-hour with sources, normalised per GB of memory and per TB/s of bandwidth, and what they leave out.