QuData

Rent H100 by the Hour by QuData Ai

Explore the specifications and benchmarks of the H100. Compare hardware configurations and performance metrics to maximize your cutting-edge AI and HPC workloads with the revolutionary GPU computing architecture.

GPU Description

Explore the specifications and benchmarks of the H100. Compare hardware configurations and performance metrics to maximize your cutting-edge AI and HPC workloads with the revolutionary GPU computing architecture.

GPU Characteristics

PCIe 5.0H100
GPU connectionPCIe
GPU model nameNVIDIA H100
Driver version535.183.01
GPU memory80GB HBM3
Power Limit (W)350.0
GPU temperature (°C)52
GPU Clock Speed (MHz)1980
Memory Clock Speed (MHz)1593
Memory bandwidth2039 GB/s

GPU Configuration

CUDA cores14592
Tensor core456 (4th generation)
Support for FP16/INT8Yes
ArchitectureHopper
VirtualizationNVIDIA MIG

Benchmarks

ffmpeg~6ms
Coremark (iterations per second)~120,000-140,000
Llama2 Inference (tokens per second)~280-350
Tensorflow Mixed Training~8.5-11.2
Stable Diffusion (images/min)~85-105
Mixed Precision Training~15.8-20.5x

Providers with H100