Loading...
Loading...
80GB HBM3 | 3.35TB/s Bandwidth | Hopper | 700W TDP | NVLink 4.0
The NVIDIA H100 SXM5 is the industry-standard accelerator for AI training and inference, featuring 80GB HBM3 with 3.35TB/s bandwidth. With NVLink 4.0 interconnect and fourth-gen Tensor Cores, it delivers breakthrough performance for transformer models and large-scale HPC.
Architecture
Hopper (GH100)
Memory
80 GB HBM3
Memory Bandwidth
3.35 TB/s
Full hardware specifications for the NVIDIA H100 SXM5
Architecture
Hopper (GH100)
Memory
80 GB HBM3
Memory Bandwidth
3.35 TB/s
TDP
700W
Interface
NVLink 4.0 / SXM5
FP16 Performance
989 TFLOPS
141GB HBM3e | 4.8TB/s Bandwidth | Hopper | 700W TDP | PCIe 5.0
Hopper (GH200)
192GB HBM3e | 8TB/s Bandwidth | Blackwell | 1000W TDP | NVLink 5.0
Blackwell (GB200)
48GB GDDR6 | 864GB/s Bandwidth | Ada Lovelace | 350W TDP | PCIe 4.0 x16
Ada Lovelace (AD102)
80GB HBM2e | 2TB/s Bandwidth | Ampere | 400W TDP | NVLink 3.0
Ampere (GA100)