CAPACITY PLANNING
GPU cluster profiles
for AI capacity planning.
Compare H100 SXM and H200 SXM reference profiles, then share the scale, region, networking and workload you need. Capacity, pricing and deployment scope require confirmation.
ACCELERATOR PROFILES
Start with the silicon.
Manufacturer specifications for reference. These profiles are starting points for a sourcing discussion, not live inventory.
| GPU | Memory per GPU | Memory bandwidth | Designed for | |
|---|---|---|---|---|
| NVIDIA H200 SXM | 141 GB HBM3e | 4.8 TB/s | Memory-intensive AI & HPC | Specifications |
| NVIDIA H100 SXM | 80 GB HBM3 | 3.35 TB/s | AI training & inference | Specifications |
LET’S SHAPE YOUR CLUSTER
Your workload sets the direction.
YOUR CAPACITY BRIEF
8 GPUs ×
H200 SXM
- Memory / GPU
- 141 GB HBM3e
- Workload
- AI training
- Region
- Discuss APAC options
This prepares a request. Capacity, locations and pricing are subject to confirmation.
LET’S CONNECT
Need a different configuration?
Share your requirements. We’ll start with your workload, not a preset.