Hetzner Cloud has introduced a new generation of shared vCPU instances, creating a simpler and more consistent structure across its server families. While the update may look confusing at first, it becomes clear once you see how the lineup is organized.
Hetzner now groups its shared-resource servers into two main categories:
- CX (Cost-Optimized x86 Cloud Servers) – designed for affordability and predictable, light workloads.
- CPX (Regular Performance AMD Cloud Servers) – focused on balanced performance and efficiency on modern AMD hardware.
These replace the previous CX and CPX generations, which are being phased out. The new lineup gives users a straightforward choice between cost efficiency and performance, depending on workload sensitivity.
CX Gen3: cost-optimized and reliable
The third generation of CX instances combines older Intel and AMD hardware from previous CX and CPX lines. It is a practical option for users who value cost efficiency over peak compute performance.
These servers are well suited for small applications, development environments, and low-intensity workloads such as blogs, forums, or lightweight backend services.
| Name | vCPUs | RAM | SSD | Architecture |
|---|---|---|---|---|
| CX23 | 2 | 4 GB | 40 GB | Intel® / AMD |
| CX33 | 4 | 8 GB | 80 GB | Intel® / AMD |
| CX43 | 8 | 16 GB | 160 GB | Intel® / AMD |
| CX53 | 16 | 32 GB | 320 GB | Intel® / AMD |
The CX family maintains Hetzner’s straightforward model while adding more flexibility. You can scale instances directly through the Hetzner Console or API and pay only for the resources you use.
CPX Gen2: regular performance, new AMD hardware
The second generation of CPX servers transitions fully to AMD Genoa CPUs, offering substantial improvements in both single-core and multi-core performance compared to earlier models.
If your workload requires consistent compute performance, such as web applications, API backends, or production systems, CPX Gen2 provides an ideal balance.
| Name | vCPUs | RAM | SSD | Architecture |
|---|---|---|---|---|
| CPX22 | 2 | 4 GB | 80 GB | AMD |
| CPX32 | 4 | 8 GB | 160 GB | AMD |
| CPX42 | 8 | 16 GB | 320 GB | AMD |
| CPX52 | 12 | 24 GB | 480 GB | AMD |
| CPX62 | 16 | 32 GB | 640 GB | AMD |
According to benchmarks shared by Sparecores, the new CPX series delivers up to five times better performance per euro than the previous generation, representing a major step forward for shared vCPU performance.
Hetzner Cloud server lineup update
This update marks a significant improvement in how Hetzner organizes its cloud server portfolio. By clearly separating cost-optimized and performance-optimized shared instances, Hetzner makes it easier to select the right server type while keeping pricing competitive.
With this release, the previous CX and CPX generations are deprecated. New instances of these older types can no longer be created through the Hetzner Console, but they remain available via Cloudfleet and automation tools such as Terraform. Existing servers will continue to operate as before.
The CAX (ARM-based) and CCX (dedicated vCPU) families remain unchanged, each serving a distinct role: CAX for energy-efficient ARM workloads, and CCX for compute-heavy production environments where dedicated CPU resources are required.
Billing remains consistent. You can choose hourly or monthly payment, with no minimum contract terms. Each plan includes generous traffic limits and an IPv4 address by default.
For more information, see the official Hetzner announcement or review current pricing and availability at hetzner.com/cloud.
Get started with Cloudfleet and Hetzner Cloud
Hetzner Cloud continues to be one of the most efficient and developer-friendly platforms in Europe. With this update, it becomes even more capable and cost-effective.
Start using Cloudfleet’s Hetzner Kubernetes integration to simplify node provisioning and scale clusters easily. Whether optimizing for cost or performance, this integration helps you run Kubernetes more efficiently.
Stay tuned for future updates as Cloudfleet expands its capabilities and deepens integration with leading cloud providers.
