All GPUs
DatacenterNone available

H100 SXM

NVIDIA H100 SXM

Industry-leading GPU for AI training with 80GB HBM3 and NVLink support.

Specifications

VRAM80 GB
Memory TypeHBM3
Bandwidth3350 GB/s
CUDA Cores16,896
Tensor Cores528
FP32 Performance67 TFLOPS
FP16 Performance134 TFLOPS
TDP700 W
ArchitectureHopper
GPU ChipGH100

Best For

LLM TrainingLarge Model InferenceHPC
Price per hour
$1.98
Cost Estimate
per hour$1.98
per day$47.52
per week$332.64
per month$1425.60