Why IBM Cloud for GPUs
A graphics processing unit (GPU) is “extra brain power” the CPU lacks. Choosing IBM Cloud® for your GPU requirements gives you direct access to one of the most flexible server-selection processes in the industry, seamless integration with your IBM Cloud architecture, APIs and applications, and a globally distributed network of data centers.
IBM Cloud Bare Metal Servers with GPUs perform better on 5 TensorFlow ML models than AWS servers.
We offer bare metal and virtual server GPUs. Google Cloud offers virtual server instances, only.
Like Google Cloud, Alibaba Cloud offers GPU options on virtual machines, only.
NVIDIA T4 Graphics Card / 20 cores / 2.20 GHz / 32 GB RAM / 20 TB bandwidth¹ / Starts at: USD 819/mo
NVIDIA T4 Graphics Card / 32 cores / 2.30 GHz / 32 GB RAM / 20 TB bandwidth¹ / Starts at: USD 934/mo
NVIDIA T4 Graphics Card / 40 cores / 2.50 GHz / 32 GB RAM / 20 TB bandwidth¹ / Starts at: USD 1,704/mo
8 vCPU / 64 GiB RAM / 1 x V100 GPU / Starts at: USD 2.49/hr
16 vCPU / 128 GiB / RAM 2 x V100 GPU / Starts at: USD 4.99/hr
32 vCPU / 256 GiB / RAM 2 x V100 GPU / Starts at: USD 5.98/hr
8 vCPU / 60 GB RAM / 1 x P100 GPU / Starts at: USD 1.95/hr
8 vCPU / 20 GB RAM / 1 x V100 GPU / Starts at: USD 3.06/hr
8 vCPU / 60 GB RAM / 1 x V100 GPU / Starts at: USD 2,233/mo
Footnote
¹ 20 TB bandwidth included in US, Canada and EU data centers; 5 TB bandwidth included in all other data centers. New prices and offers may not be combined with any other current or future discounts.