All systems operational
DocsStatusSupport

GPU Cloud for AI/ML

Deploy high-performance GPU instances in minutes. From development to production, scale with confidence.

Available GPUs
4,200+
Regions
7
Deploy Time
7-10 min
Uptime SLA
99.99%
12ms latency

14 GPUs available

NVIDIA T4

2048 available

Budget-friendly inference GPU

PCIeMulti-GPU99.5% SLASOC2
Memory
16 GB
FP16
65 TFLOPS
Bandwidth
300 GB/s
$0.35/hr
~$255/mo

NVIDIA A10G

1024 available

Cost-effective GPU for inference workloads

PCIeMulti-GPU99.5% SLASOC2
Memory
24 GB
FP16
125 TFLOPS
Bandwidth
600 GB/s
$0.65/hr
~$475/mo

NVIDIA L4

640 available

Efficient inference GPU for production

PCIeMulti-GPU99.9% SLASOC2
Memory
24 GB
FP16
121 TFLOPS
Bandwidth
300 GB/s
$0.85/hr
~$621/mo

NVIDIA RTX 4090

512 available

High-performance GPU for development

PCIe99.5% SLA
Memory
24 GB
FP16
165 TFLOPS
Bandwidth
1.0 TB/s
$0.95/hr
~$694/mo

NVIDIA RTX 6000 Ada

Enterprise256 available

Professional workstation GPU

PCIeMulti-GPU99.5% SLASOC2
Memory
48 GB
FP16
183 TFLOPS
Bandwidth
960 GB/s
$1.45/hr
~$1059/mo

NVIDIA L40S

Enterprise384 available

Versatile GPU for inference and video processing

PCIeMulti-GPU99.9% SLASOC2
Memory
48 GB
FP16
362 TFLOPS
Bandwidth
864 GB/s
$1.80/hr
~$1314/mo

NVIDIA A100 PCIe 40GB

Enterprise320 available

Cost-effective A100 for smaller workloads

PCIeMulti-GPU99.9% SLASOC2
Memory
40 GB
FP16
624 TFLOPS
Bandwidth
1.6 TB/s
$2.20/hr
~$1606/mo

NVIDIA A100 PCIe

Enterprise256 available

High-performance GPU for AI training and HPC

PCIeMulti-GPU99.9% SLASOC2
Memory
80 GB
FP16
624 TFLOPS
Bandwidth
2.0 TB/s
$2.80/hr
~$2044/mo

NVIDIA A100 SXM

Enterprise192 available

Enterprise GPU for AI training and HPC

NVLinkMulti-GPU99.9% SLASOC2, HIPAA
Memory
80 GB
FP16
624 TFLOPS
Bandwidth
2.0 TB/s
$3.20/hr
~$2336/mo

NVIDIA H100 PCIe

Flagship96 available

High-performance GPU with PCIe flexibility

PCIeMulti-GPU99.95% SLASOC2, HIPAA
Memory
80 GB
FP16
1,513 TFLOPS
Bandwidth
2.0 TB/s
$3.80/hr
~$2774/mo

NVIDIA H100 SXM

Flagship128 available

Flagship GPU for large-scale AI training

NVLinkMulti-GPU99.95% SLASOC2, HIPAA
Memory
80 GB
FP16
1,979 TFLOPS
Bandwidth
3.35 TB/s
$4.50/hr
~$3285/mo

NVIDIA H200

Flagship64 available

Enhanced Hopper with expanded HBM3e memory

NVLinkMulti-GPU99.95% SLASOC2, HIPAA
Memory
141 GB
FP16
1,979 TFLOPS
Bandwidth
4.8 TB/s
$6.80/hr
~$4964/mo

NVIDIA B100

Flagship48 available

High-performance Blackwell for AI workloads

NVLinkMulti-GPU99.95% SLASOC2, HIPAA
Memory
80 GB
FP16
3,500 TFLOPS
Bandwidth
6.0 TB/s
$8.50/hr
~$6205/mo

NVIDIA B200

Flagship24 available

Next-gen flagship for frontier AI models

NVLinkMulti-GPU99.99% SLASOC2, HIPAA, GDPR
Memory
192 GB
FP16
4,500 TFLOPS
Bandwidth
8.0 TB/s
$12.50/hr
~$9125/mo

Need more power?

Contact us for custom configurations, reserved capacity, and enterprise support.

Contact Sales