H100
The reference is H100 SXM 80 GB. H100 NVL uses a different memory configuration; an offer must identify the exact variant.
Explore H100 →GLOBALATLAS AI · H100 vs B200
H100 vs B200
Compare H100 and B200 memory, architecture and deployment considerations using manufacturer specifications.
| Parameter | H100 | B200 |
|---|---|---|
| GPU memory | 80 GB | 180 GB |
| Memory technology | HBM3 | HBM3e |
| Architecture | Hopper | Blackwell |
| Workload profile | AI training, inference and accelerated computing | Training and inference with high memory demand |
| Reference variant | The reference is H100 SXM 80 GB. H100 NVL uses a different memory configuration; an offer must identify the exact variant. | This page describes the 180 GB HGX B200 configuration. Other platform variants must be checked against their exact specification. |
| Price | Price on requestConfiguration and region to be confirmed | Price on requestConfiguration and region to be confirmed |
Which configuration fits your workload?
B200
Compare B200 with H100 when the working set or topology is changing. Benchmark the same model and precision on both systems.
Explore B200 →Compare the same workload
Use the same model or scene, precision, batch size and software version. Record throughput, latency, peak memory and total server cost. This table contains hardware specifications, not measured application performance.