New Arrival

NVIDIA H200 SXM5
ALIVE ON CLOUD

Take your AI capabilities to the next level.
Early and instant access to the H200 GPUs starting at $0.99/h.
NVIDIA H200 SXM5 ALIVE ON CLOUD
1.9x
Faster Llama2 70B Inference
1.6x
Faster GPT-3 175B Inference
110x
Faster High-Performance Computing

Compared to NVIDIA H100 NVL, H200 excels in all benchmark tests.

Performance

Enjoy Unmatched GPU Performance

NVIDIA H200 GPU provides outstanding performance for AI training, fine-tuning, and inference, with
enhanced memory capacity and ultra-high bandwidth.

High-Performance LLM Inference

The H200 boosts inference performance, offering superior speed and efficiency for large-scale AI applications like Llama2.

H100
H200

Supercharge High-Performance Computing

The H200’s superior memory bandwidth ensures efficient data handling, boosting performance for memory-intensive HPC applications like AI and simulations.

H100
H200

Reduce Energy and TCO

H200 delivers breakthrough performance and energy efficiency, driving AI and supercomputing advancements with a lower TCO.

H100
H200
Data information from NVIDIA H200 page:https://www.nvidia.com/en-sg/data-center/h200/?ncid=no-ncid
Technical Specs

NVIDIA H200 SXM5

CPU2 x Intel Xeon Platinum 8580
GPU8 x NVIDIA H200-SXM5-141GB
Memory32 x DDR5 5600MHz 64GB
System disk2 x 960GB NVMe SSD
Data disk8 x 7.68TB NVMe SSD
Compute Network8 x 400Gb/s InfiniBand
Storage Network200Gb/s
In-Band Network25Gb/s
See Full Pricing →
NVIDIA H200 NVL Tensor Core GPU 8 x 141GB ON DEMAND compute

Flexible Pricing Options

Best Price Guaranteed. Choose the pricing model that best fits your workload requirements. Learn More >

On-Demand

$0.99 /hr
Per H200 GPU
  • Instant deployment
  • Pay per second billing
  • No long-term commitment
  • Global availability
  • 99% uptime SLA

Reserved Cluster

Contact Sales
for customized requirements
  • Bare Metal or Cloud Access
  • Priority access
  • Custom configurations
  • Dedicated support
  • 99% uptime SLA

Dynamic Pricing

from $1.99 /hr
Per H200 GPU
  • Up to 36% savings
  • Flexible workloads
  • Pay per second billing
  • No long-term commitment
  • Global availability

Ready to Power your AI Workloads?

Join thousands of developers and researchers who are building the future with our HPC-AI platform.