GPU Servers

Ephemeral, Scalable GPU servers

Cato’s bare metal GPU servers feature graphics processing units (GPUs) specifically designed to accelerate complex computing tasks such as machine learning, scientific simulations, and virtual reality. They are ideal for workloads requiring massive parallel processing power, reducing processing times from days or weeks to hours or even minutes. These servers are available at considerably lower costs than at the big cloud providers for equivalent performance.


g1.p2l.1nv g1.p2xl.2nv8
Processor: 2 x Intel Xeon E5-2680v4 2 x Intel Xeon E5-2680v4
vCores: 56 56
Processor Speed: 2.4-3.3Ghz 2.4-3.3Ghz
Monthly †: $657 $5,256
Hourly ‡: $1.50 $12.00  (8 x $1.50)
SLA: 99.5% 99.5%
Memory: 256GiB 512GiB
Local Storage: 1 x 512GB SSD 1 x 512GB SSD
Attached Storage:
Network Speed: 10Gbps 10Gbps
Data Transfer ✲: $0.0025/GiB $0.0025/GiB
GPU: 1x Nvidia T4 8x Nvidia V100
Cuda Cores: 2,560 40,960
Tensor Cores: 320 5,120
Subject to availiability; subject to change; prices shown in USD; † shown with annual prepayment; ‡ calculated as a 30 day month; ✲ for egress/ingress from Pod.
Cato Digital

Ready to get started?

View available application or storage servers. Get Started
Cato Digital ™, and © Cato Digital, inc  | Terms | Privacy