Product Compare

NVIDIA H100 SXM5 vs Intel Data Center GPU Max 1350

GPU GPU
Graphics Card
Release Date Mar 2022 Jan 2023
Generation Tesla Hopper Data Center GPU
Type AI GPU Professional
Bus Interface PCIe 5.0 x16 PCIe 5.0 x16
Clock Speeds
Base Clock 1590 MHz 750 MHz
Boost Clock 1980 MHz 1550 MHz
Memory Clock 1313 MHz 1200 MHz
Memory
Memory Size 80GB 96GB
Memory Type HBM3 HBM2e
Memory Bus 5120bit 8192bit
Bandwidth 3350GB/s 2458GB/s
Render Config
SM Count 132
Shading Units 16896 14336
TMUs 528 896
ROPs 24 0
Tensor Cores 528 896
L1 Cache 256 KB (per SM) 64 KB (per EU)
L2 Cache 50 MB 408 MB
RT Cores 112
Theoretical Performance
Pixel Rate 47.52 GPixel/s 0 MPixel/s
Texture Rate 1045 GTexel/s 1389 GTexel/s
FP16 (half) 267.6 TFLOPS 44.44 TFLOPS
FP32 (float) 66.91 TFLOPS 44.44 TFLOPS
FP64 (double) 33.45 TFLOPS 44.44 TFLOPS
Graphics Processor
GPU Name GH100 Ponte Vecchio
Architecture Hopper Generation 12.5
Foundry TSMC Intel
Process Size 4 nm 10 nm
Transistors 80 billion 100 billion
Die Size 814 mm² 1280 mm²
Board Design
TDP 700W 450W
Suggested PSU 1100 W 850 W
Outputs No outputs No outputs
Power Connectors 8-pin EPS
Graphics Features
DirectX N/A 12 (12_1)
OpenGL N/A 4.6
OpenCL 3.0 3.0
Vulkan N/A N/A
CUDA 9.0
Shader Model N/A 6.6