NVIDIA H200 Tensor Core GPU
The NVIDIA H200 supercharges generative AI and high-performance computing (HPC) workloads with game-changing performance and memory capabilities.
Key Facts
1.9X FASTER
Llama2 70B Inference
The H200 provides up to 1.9X faster inference on Llama2 70B than the H100.
141 GB
HBM3e Memory
The first GPU to offer 141 gigabytes (GB) of HBM3e memory.
4.8 TB/s
Memory Bandwidth
1.4X more memory bandwidth than the previous H100 GPUs.
110X FASTER
HPC Performance
H200’s higher memory bandwidth leads to 110X faster performance compared to CPUs.
Our Nodes
Based on the NVIDIA Hopper™ architecture, the NVIDIA H200 is the first GPU to offer 141 gigabytes (GB) of HBM3e memory at 4.8 terabytes per second (TB/s) —that’s nearly double the capacity of the NVIDIA H100 Tensor Core GPU with 1.4X more memory bandwidth.
