All GPUs
NVIDIA · Hopper
NVIDIA HGX H200
Hopper with 141 GB of faster HBM3e per GPU, for larger models and longer contexts than H100.
- On-demand
- 2 years
- 3 years
Specifications
- Architecture
- Hopper
- GPU memory
- 141 GB HBM3e
- Memory bandwidth
- 4.8 TB/s
- GPU-to-GPU
- NVLink 4 · 900 GB/s
- Form factor
- SXM (HGX)
Best for
- Large-model inference and long contexts
- Training and fine-tuning
- Upgrading H100 workloads in place