Loading the catalog…
H100 SXM (NVIDIA)
Blurred values are subscriber-only. Sign in to start a free trial — 2 full pages a day — or subscribe for unlimited access.
| Memory Bandwidth Tb S | 3.35 |
|---|---|
| Process Node | TSMC 4N |
| Fp8 Tflops Dense | 1979 |
| Interconnect | NVIDIA NVLink: 900GB/s, PCIe Gen5: 128GB/s |
| Fp32 Tflops | 67 |
| Tdp W | 700 |
| Gpu Memory Gb | 80 |
| Accelerator Form Factor | SXM |
| Architecture | Hopper |
| Fp16 Bf16 Tflops Dense | 989.5 |
| Fp16 Bf16 Tflops Sparse | 1979 |
| Int8 Tops Dense | 1979 |
| Fp64 Tflops | |
| Fp8 Tflops Sparse | |
| Int8 Tops Sparse | |
| Media Engines | |
| Product Name | |
| Tf32 Tflops Dense | |
| Tf32 Tflops Sparse |
| SKU | H100 SXM |
|---|
| Product | |
|---|---|
| BFLOAT16 Tensor Core* | |
| INT8 Tensor Core* | |
| Multi-Instance GPUs | |
| Server Options | |
| NVIDIA Enterprise | |
| Transistors | |
| NVLink Switch System bandwidth | |
| Multi-Instance GPU generation |
