DatacenterNone available
Rent an H200
NVIDIA H200
The most powerful GPU for AI training and large language models with 141GB HBM3e memory.
Specifications
VRAM141 GB
Memory TypeHBM3e
Bandwidth4800 GB/s
CUDA Cores16,896
Tensor Cores528
FP32 Performance67 TFLOPS
FP16 Performance134 TFLOPS
TDP700 W
ArchitectureHopper
GPU ChipGH100
Best for
Large LLM TrainingMulti-model InferenceResearch
Price per hour
$4.20
Cost estimate
per hour$4.20
per day$100.80
per week$705.60
per month$3024.00
Need H200 servers for months rather than hours? Reserved H200 capacity from data centers.