Purchase Orders & Tax-Exempt
For schools, universities, government agencies, and qualified businesses — better pricing, NET-30 terms, and tax-exempt orders welcome.
Request a quote →NVIDIA H200 141GB HBM3e AI Data Center GPU
The NVIDIA H200 GPU supercharges generative AI and high-performance computing workloads with advanced Hopper architecture capabilities. Featuring 141GB of ultrafast HBM3e memory delivering 4.8TB/s bandwidth, it significantly accelerates large language model inference and complex scientific simulations.
Engineered for enterprise scale, the H200 provides enhanced energy efficiency and lower total cost of ownership across SXM and PCIe configurations. Its robust architecture delivers high-throughput computing performance to meet the demands of modern enterprise AI infrastructure.
| Brand | NVIDIA |
| Model | H200 |
| Architecture | NVIDIA Hopper |
| GPU Memory | 141GB HBM3e |
| Memory Bandwidth | 4.8TB/s |
| Form Factor Options | SXM / PCIe Dual-Slot |
| FP8 Tensor Core Performance | Up to 3,958 TFLOPS |
| FP16 Tensor Core Performance | Up to 1,979 TFLOPS |
| TF32 Tensor Core Performance | Up to 989 TFLOPS |
| FP64 Performance | Up to 34 TFLOPS |
| Interconnect | NVIDIA NVLink (900GB/s), PCIe Gen5 (128GB/s) |
| Confidential Computing | Supported |
| Multi-Instance GPU (MIG) | Up to 7 MIGs |
| Estimated Shipping Weight | 4.5 lb |
3999900
52534482043160
NVIDIA H200 141GB HBM3e AI Data Center GPU
10381829865752
NVIDIA H200 141GB HBM3e AI Data Center GPU
