NWS
GPUs

GPU power for AI, machine learning and high-performance computing

Start free of charge
GPUsAdvantagesUse CasesNext stepsFAQ

Our available GPUs

NVIDIA A10

  • For small models

NVIDIA A40

  • For medium-sized models

NVIDIA RTX Pro 6000

  • For large models
wdt_ID Merkmal NVIDIA A10 NVIDIA A40 NVIDIA RTX Pro 6000
14 Architektur Ampere-Architektur Ampere-Architektur Blackwell Architektur
15 Speicher 24GB GDDR6 VRAM 48GB GDDR6 ECC VRAM 96GB GDDR7 ECC VRAM
16 Speicherbandbreite 600 GB/s 798 GB/s 1597 GB/s
17 Rechenleistung FP32 31 teraFLOPS FP32 37 teraFLOPS FP32 120 teraFLOPS, FP4 AI Compute (Peak) 4 petaFLOPS
18 Einsatzbereiche AI-Inference, Datenanalyse Simulationen, Deep Learning & AI-Inference komplexe Simulationen, Deep Learning, AI-Finetuning & AI-Inference
19 AI-Modelle z.B. GPT-OSS-20B oder Qwen3-8B z.B. Ministral-3-14B oder Qwen3-14B z.B. GPT-OSS-120B oder Qwen3.5-27B
21 MIG (Multi-Instance-GPU) X X ✓

Your advantages

Maximum performance for demanding workloads

Test free of charge Icon

High availability

Stable infrastructure for critical applications.
High-performance GPUs ensure the optimum workload for your AI applications.

Optimized performance

Modern hardware with high computing power and scalable resources.
Security and data protection icon

Security and data protection

Our data centers in Germany are ISO-certified and we operate in compliance with the GDPR.
Managed AI solution with direct API connection.

Integration into existing workflows

Easy integration into existing cloud or Kubernetes environments.
Create an NWS ID and get started right away!
Start free of charge

Typical use cases

Artificial Intelligence & Machine Learning

High Performance Computing

Data Science & Big Data

Rendering & Visualization

FAQ

2
3
When is the use of GPUs worthwhile?
The use of GPUs is particularly worthwhile for computationally intensive tasks. These include machine learning, large data analyses, simulations and rendering processes. GPUs can process these workloads significantly faster than traditional CPU systems.
2
3
Which applications particularly benefit from GPUs?
Typical applications are

  • Training of AI and machine learning models
  • Data analysis and big data processing
  • 3D rendering and visualization
  • Scientific simulations
  • Video processing

2
3
Can GPU resources be scaled flexibly?
Yes, the GPU resources can be scaled flexibly. Depending on requirements, additional GPUs can be provided or workloads can be distributed across several GPUs.
2
3
Are GPUs also useful for small projects?
Even smaller projects can benefit from GPU acceleration, for example when training initial machine learning models or processing large data sets. GPUs can be used as required thanks to flexible provisioning models.
2
3
How is GPU performance measured?

The performance of a GPU is determined by various factors, including

  • Architecture of the GPU
  • Memory size and memory bandwidth
  • Computing power

2
3
Who can I contact if I have problems?
Please contact us by e-mail at nws@netways.de or via our contact form.