NVIDIA H200 141GB HBM3e GPU. 4.8 TB/s memory bandwidth, 141GB HBM3e, 700W SXM. Hopper architecture — the ultimate accelerator for large-scale AI training and inference.
H200 141GB
Hopper
141 GB HBM3e
4.8 TB/s
34 TFLOPS
989 TFLOPS
1,979 TFLOPS
SXM5 / PCIe 5.0
700W
SXM
~$25,000-35,000
NVIDIA H200 with 141GB HBM3e is purpose-built for the largest AI models. The massive memory capacity enables training trillion-parameter models, running massive LLM inference batches, and pushing the boundaries of scientific computing. Essential for frontier AI labs and hyperscale cloud providers.
Need NVIDIA H200 141GB?
Enterprise hardware — configured, tested, deployed. Contact us for pricing and availability.
Request Quote