GPU Servers for AI, Rendering, and Simulation
Compute power for demanding workloads, on HPE enterprise hardware
- NVIDIA RTX PRO 6000 Blackwell with 96 GB
- AMD EPYC with 32 cores and 256 GB RAM
- Operated in certified data centers in Germany
GPU Server
Compute power you can trust: operated in Germany
AI workloads often process sensitive data, such as customer information, documents, or trade secrets. Beyond performance, what matters most is where and how your infrastructure is operated. The GPU Server runs in certified data centers in Germany. Your data stays within the European legal jurisdiction and is subject to German and European data protection law. The foundation is HPE enterprise hardware designed for continuous operation. The result is infrastructure you can rely on.
Full GPU performance
96 GB of GPU memory for large models, datasets, and complex 3D scenes.
Enterprise hardware
HPE servers with AMD EPYC, designed for continuous operation.
Control over your data
Run your models in an environment that you control yourself.
GPU Server price list
The GPU Server is offered as a fixed configuration: GPU, processor, and memory are clearly defined. You know exactly what performance you are getting and can plan your project with confidence.
- Hosting in Germany
- Discounts for longer terms
- 24/7 support
GPU Server
From 1,900
€
per Month
-
GPU: NVIDIA RTX PRO 6000 Blackwell Server Edition, 96 GB
-
CPU: AMD EPYC 9355P, 32 cores, 3.55 GHz
-
RAM : 256 GB
ab November
What the GPU Server is suited for
A dedicated GPU with 96 GB of memory covers a wide range of scenarios, from language models to fluid dynamics simulation.
Train, customize, and run language models
Run open-source language models in your own environment, adapt existing models to your data through fine-tuning, or use inference for chatbots, assistants, and text analysis. The large GPU memory has room for models that no longer run on standard GPUs.
Faster Graphics, Less Waiting
Graphics-intensive tasks without the wait
Render 3D scenes, process video effects, or create architectural visualizations. The GPU accelerates compute-intensive graphics jobs that would take hours on workstations.
Große Datenmengen, schnelle Ergebnisse
Analyze large volumes of data faster
Run scientific simulations, forecasting models, and data analyses with GPU acceleration. Get results faster, even with very large datasets.
NEW
Ready for the November launch?
Be the first to know when sales begin: by registering, you will be among the first to learn when the plusserver GPU Server becomes available to order.
Get your GPU Server in three steps
Register
Enter your contact details in the registration form. It only takes a few minutes.
Wait for the launch
As soon as the GPU Server is available to order in November, we will notify you by email.
Order and get started
Choose your configuration and start your projects.
Security and compliance you can demonstrate
The GPU Server runs in certified data centers in Germany. plusserver operates its infrastructure according to recognized standards and has it audited externally on a regular basis.
Frequently asked questions about the GPU Server (FAQ)
A GPU Server is a server with a graphics processing unit (GPU) that performs a very large number of calculations simultaneously. This makes it significantly faster than a CPU-only server, for example when training and running AI models, rendering, and running simulations.
For AI training, fine-tuning, and inference as well as for rendering, simulation, and data analysis.
Both. With 96 GB of GPU memory, the GPU Server is particularly well suited for inference and for fine-tuning existing models.
Yes. You can run open-source language models on the GPU Server in your own environment. Your data and prompts remain under your control.
That depends on the architecture, precision, and context length. As a rough guideline, with 96 GB you can run many models of up to about 70 billion parameters at reduced precision (e.g., FP8).
The GPU Server runs in certified data centers in Germany. Your data stays within the European legal jurisdiction and is subject to German and European data protection law.