AI Compute Services
High-performance Compute as a Service designed for AI-Driven Businesses, AI Developers, and AI Researchers.
Ideal for self-managed AI Training, Inference, and large-scale data processing.
NVIDIA GH200
1
96GB
72
480 GB
2x 3.8TB NVMe
IB 400G
$3.87 / GPU
NVIDIA H100 SXM
8
80 GB
208
1024 GB
6x 3.8TB NVMe
IB 3200G
$2.10 / GPU
NVIDIA A100 SXM
8
80 GB
208
1024 GB
6x 3.8TB NVMe
IB 1600G
$1.30 / GPU
Intel Gaudi 2
8
96 GB
160
1024 GB
4x 7.6TB NVMe
RoCE 2400G
$1.25 / GPU
NVIDIA A100 PCIe
4
40 GB
64
512 GB
2x 3.8TB NVMe
-
$1.15 / GPU
NVIDIA A100 SXM
8
40 GB
128
1024 GB
4x 3.8TB NVMe
IB 800G
$1.15 / GPU
NVIDIA A40 PCIe
4
48 GB
128
512 GB
2x 3.8TB NVMe
-
$0.65 / GPU
Platform
GPUs
GPU VRAM
vCPUs
Memory
Local Storage
Interconnect
Price/Hour
Access instances immediately with no term commitments. Available in Bare Metal and Virtualization with 1x, 2x, 4x, and 8x multi-GPU configurations.
On-demand GPU instances
Select GPU configuration:
Intel Xeon Gen3 (Ice Lake)
8
32 GB
$0.31 / hour
Intel Xeon Gen3 (Ice Lake)
16
64 GB
$0.62 / hour
Intel Xeon Gen3 (Ice Lake)
32
128 GB
$1.24 / hour
AMD EPYC 3rd Gen (Milan)
8
32 GB
$0.40 / hour
AMD EPYC 3rd Gen (Milan)
16
64 GB
$0.79 / hour
AMD EPYC 3rd Gen (Milan)
32
128 GB
$1.59 / hour
Platform
vCPUs
Memory
Price/Hour
Use CPU virtual machines for data processing, or GPU orchestration software like Kubernetes and SLURM.
On-demand CPU instances
8
GPUs |
Price/Hour |
$1.25 / GPU
96 GB
GPU VRAM |
160
vCPU |
Interconnect |
RoCE 2400G
Local Storage |
4x 7.6TB NVMe
1024 GB
Memory |
1
GPUs |
Price/Hour |
$3.87 / GPU
96 GB
GPU VRAM |
72
vCPU |
Interconnect |
IB 400G
Local Storage |
2x 3.8TB NVMe
480 GB
Memory |
8
GPUs |
Price/Hour |
$2.44 / GPU
80 GB
GPU VRAM |
208
vCPU |
Interconnect |
IB 3200G
Local Storage |
6x 3.8TB NVMe
1024 GB
Memory |
8
GPUs |
Price/Hour |
$1.40 / GPU
80 GB
GPU VRAM |
208
vCPU |
Interconnect |
IB 1600G
Local Storage |
6x 3.8TB NVMe
1024 GB
Memory |
8
GPUs |
Price/Hour |
$1.15 / GPU
40 GB
GPU VRAM |
128
vCPU |
Interconnect |
IB 800G
Local Storage |
4x 3.8TB NVMe
1024 GB
Memory |
4
GPUs |
Price/Hour |
$1.15 / GPU
40 GB
GPU VRAM |
64
vCPU |
Interconnect |
-
Local Storage |
2x 3.8TB NVMe
512 GB
Memory |
4
GPUs |
Price/Hour |
$1.00 / GPU
48 GB
GPU VRAM |
128
vCPU |
Interconnect |
-
Local Storage |
2x 3.8TB NVMe
512 GB
Memory |
• VM instances available with 1X, 2X, 4X and 8X GPUs
• Virtualization or Bare Metal options
• No-term commitments required
On-demand GPU instances
AI Compute Services
High-performance Compute as a Service designed for AI-Driven Businesses, AI Developers, and AI Researchers.
Ideal for self-managed AI Training, Inference, and large-scale data processing.
Intel Xeon Gen3 (Ice Lake)
8
Price |
vCPU |
32 GB
Memory |
$0.31 / hour
Intel Xeon Gen3 (Ice Lake)
16
Price |
vCPU |
64 GB
Memory |
$0.62 / hour
Intel Xeon Gen3 (Ice Lake)
32
Price |
vCPU |
128 GB
Memory |
$1.24 / hour
AMD EPYC 3rd Gen (Milan)
8
Price |
vCPU |
32 GB
Memory |
$0.40 / hour
AMD EPYC 3rd Gen (Milan)
16
Price |
vCPU |
64 GB
Memory |
$0.79 / hour
AMD EPYC 3rd Gen (Milan)
32
Price |
vCPU |
128 GB
Memory |
$1.59 / hour
Use CPU virtual machines for data processing, or GPU orchestration software like Kubernetes and SLURM.
On-demand CPU instances
Intel Xeon Gen3 (Ice Lake)
8
32 GB
$0.31 / hour
Price/Hour |
Memory |
vCPUs |
Intel Xeon Gen3 (Ice Lake)
16
64 GB
$0.62 / hour
Price/Hour |
Memory |
vCPUs |
Intel Xeon Gen3 (Ice Lake)
32
128 GB
$1.24 / hour
Price/Hour |
Memory |
vCPUs |
AMD EPYC 3rd Gen (Milan)
8
32 GB
$0.40 / hour
Price/Hour |
Memory |
vCPUs |
AMD EPYC 3rd Gen (Milan)
16
64 GB
$0.79 / hour
Price/Hour |
Memory |
vCPUs |
AMD EPYC 3rd Gen (Milan)
32
128 GB
$1.59 / hour
Price/Hour |
Memory |
vCPUs |
Use CPU virtual machines for data processing, or GPU orchestration software like Kubernetes and SLURM.
Storage and Networking
File and block storage with no commitments or limits. Networked storage uses distributed data redundancy for high availability workloads.
Storage Services
Price
Network Services
Price
General purpose
$0.03 per GB/mo
High performance NVMe SSD
$0.08 per GB/mo
Internet ingress and egress
Free
Public IPv4 addresses
Free
VPN gateways
Free
VPCs
Free
On-demand CPU instances
Pre-selling now!
Contact our team for priority access to Next Generation AI Compute Architectures.
NVIDIA Blackwell GPUs coming soon!
Platform
GPUs
GPU VRAM
vCPUs
Memory
Local Storage
Interconnect
NVIDIA GB200
2
384 GB
72 Arm®
2048 GB
6x 3.8TB NVMe
RoCE or IB 3200G
NVIDIA B200 SXM
8
180 GB
224
2048 GB
6x 3.8TB NVMe
RoCE or IB 3200G
Intel Gaudi 3
8
128 GB
224
2048 GB
4x 7.6TB NVMe
RoCE 4800G
AMD Mi300X
8
192 GB
224
2048 GB
6x 3.8TB NVMe
RoCE 3200G
Pre-selling now!
Contact our team for priority access to Next Generation AI Compute Architectures.
NVIDIA Blackwell GPUs coming soon!
2
GPUs |
384 GB
GPU VRAM |
72 Arm®
vCPU |
Interconnect |
RoCE or IB 3200G
Local Storage |
6x 3.8TB NVMe
2048 GB
Memory |
8
GPUs |
180 GB
GPU VRAM |
224
vCPU |
Interconnect |
RoCE or IB 3200G
Local Storage |
6x 3.8TB NVMe
2048 GB
Memory |
8
GPUs |
128 GB
GPU VRAM |
224
vCPU |
Interconnect |
RoCE 4800G
Local Storage |
4x 7.6TB NVMe
2048 GB
Memory |
8
GPUs |
192 GB
GPU VRAM |
224
vCPU |
Interconnect |
RoCE 3200G
Local Storage |
6x 3.8TB NVMe
2048 GB
Memory |
Storage and Networking
File and block storage with no commitments or limits. Networked storage uses distributed data redundancy for high availability workloads.
Storage Services
Price
Network Services
Price
General purpose
$0.03 per GB/mo
Price |
High performance NVMe SSD
$0.08 per GB/mo
Price |
Internet ingress and egress
Free
Price |
Public IPv4 addresses
Free
Price |
VPN gateways
Free
Price |
VPCs
Free
Price |
Storage and Networking
Network Services
