NVIDIA H200 SXM vs Intel Gaudi 3 Full Specs

Summary

16,896 Shaders 1.98GHz	16,384 Shaders 438MHz
141GB HBM3e4.81TB/s	128GB HBM2e3.7TB/s
Peak·FP32·FP16 66.91 TFLOPS	Peak·FP32·FP16 -
Form Factor SXM5	Form Factor OAM Module
TDP 700W	TDP 900W

Highlights

Peak OPS

Peak OPS FP32 FP16 Tensor FP8-16 FP8-16 Tensor Sparse Tensor FP8-32 FP8-32 Tensor Sparse Tensor FP16-16 FP16-16 Tensor Sparse Tensor FP16-32 FP16-32 Tensor Sparse Tensor FP32 FP64 Tensor FP64 BF16 Tensor BF16 BF16 Tensor Sparse Tensor TF32 Tensor INT8 Memory Bandwidth TDP

H200 SXM3.96 POPSINT8 Tensor Sparse

x2.15

Gaudi 31.84 PFLOPSFP8-32 Tensor

x1

Show Top Score

Theoretical Performance

Clock Speed1980MHz UC·Base·Boost·OC	Clock Speed438MHz UC·Boost·OC
Peak OPS 3.96 POPSINT8 Tensor Sparse	Peak OPS 1.84 PFLOPSFP8-32 Tensor
Tensor FP8-16 1.98 PFLOPS FP8-16 Tensor Sparse 3.96 PFLOPS Tensor FP8-32 1.98 PFLOPS FP8-32 Tensor Sparse 3.96 PFLOPS	Tensor FP8-16 - FP8-16 Tensor Sparse - Tensor FP8-32 1.84 PFLOPS FP8-32 Tensor Sparse -
Tensor FP16-16 989.4 TFLOPS FP16-16 Tensor Sparse 1.98 PFLOPS Tensor FP16-32 989.4 TFLOPS FP16-32 Tensor Sparse 1.98 PFLOPS	Tensor FP16-16 - FP16-16 Tensor Sparse - Tensor FP16-32 459.3 TFLOPS FP16-32 Tensor Sparse -
BF16 133.8 TFLOPS Tensor BF16 989.4 TFLOPS BF16 Tensor Sparse 1.98 PFLOPS	BF16 - Tensor BF16 1.84 PFLOPS BF16 Tensor Sparse -
Tensor TF32 494.7 TFLOPS	Tensor TF32 459.3 TFLOPS
FP32 66.91 TFLOPS Tensor FP32 -	FP32 - Tensor FP32 229.6 TFLOPS
FP64 33.45 TFLOPS Tensor FP64 61.84 TFLOPS	-
Tensor INT8 1.98 POPS	-
Pixel Rate 47.5 GPixel/s	Pixel Rate -
Texture Rate 1.04 TTexel/s	Texture Rate -

Core Config

Shaders 16,896 Shaders	Shaders 16,384 Shaders
TMUs 528 TMUs	TMUs -
ROPs 24 ROPs	ROPs -
Tensor Cores 528 T-Cores	Tensor Cores 64 T-Cores
SMs 132 SMs	CUs 8 CUs

Clocks

Base Clock 1.06GHz	Base Clock -
Boost Clock 1.98GHz	Boost Clock 438MHz
Tensor Clock 1.83GHz	Tensor Clock -

Cache

L2 Cache 51.2MB shared

L2 Cache -

Memory

141GB HBM3e	128GB HBM2e
Memory Bus 6144-bit	Memory Bus 8192-bit
Memory Speed 6.3GT/s	Memory Speed 3.6GT/s
Memory Bandwidth 4.81TB/s	Memory Bandwidth 3.7TB/s
ECC -	ECC Yes
ESRAM -	ESRAM 96MBSize 12800GB/sBandwidth

Power & Thermals

TDP 700W

TDP 900W

Decoder

Decoder Model 7x NVDEC 5

Decoder Model 14x Gaudi

Card

Form Factor SXM5	Form Factor OAM Module
Cooling Passive	Cooling Passive

Chip

Manufacturer NVIDIA	Manufacturer Intel
Chip Designer NVIDIA	Chip Designer Intel
Architecture Hopper	Architecture Gaudi 3
Family Server	Family Gaudi
Branding	Branding
Codename NV180	Codename Gaudi 3
Market Segment Server	Market Segment Server
Release Date Nov 13, 2023	Release Date Apr 9, 2024

Fabrication

Foundry TSMC	Foundry TSMC
Fabrication Node 4N	Fabrication Node N5
Die Size 814mm²	Die Size -
Transistor Count 80 Billion	Transistor Count -
Transistor Density 98.28 MTr/mm²	Transistor Density -

Images

No images available

No images available

Competitors

NVIDIAH200 SXM vs GPU NVIDIAB100 NVIDIAH200 SXM vs GPU NVIDIAB200 NVIDIAH200 SXM vs GPU NVIDIAGB200 Show 12 more

IntelGaudi 3 vs GPU AMDInstinct MI250X IntelGaudi 3 vs GPU AMDInstinct MI250 IntelGaudi 3 vs GPU AMDInstinct MI300X Show 12 more