NVIDIA A16 vs NVIDIA H100 SXM Full Specs

Summary

5,120 Shaders 1.7GHz	16,896 Shaders 1.98GHz
64GB GDDR6463.4GB/s	96GB HBM33.35TB/s
Peak·FP32·FP16 17.36 TFLOPS	Peak·FP32·FP16 66.91 TFLOPS
Form Factor PCIe Card	Form Factor SXM5
TDP 250W	TDP 700W
Power Connectors 1x 8-Pin EPS	Power Connectors -

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 FP64 Tensor FP64 BF16 Tensor BF16 BF16 Tensor Sparse Tensor TF32 Tensor INT8 Tensor INT4 Memory Bandwidth TDP

A16555.4 TOPSINT4 Tensor Sparse (4x 138.9 TOPS)

x1

H100 SXM3.96 POPSINT8 Tensor Sparse

x7.13

Show Top Score

Theoretical Performance

Clock Speed1695MHz UC·Base·Boost·OC	Clock Speed1980MHz UC·Base·Boost·OC
Peak OPS 555.4 TOPSINT4 Tensor Sparse	Peak OPS 3.96 POPSINT8 Tensor Sparse
-	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 FP16-16 69.43 TFLOPS FP16-16 Tensor Sparse 138.9 TFLOPS Tensor FP16-32 34.71 TFLOPS FP16-32 Tensor Sparse 69.43 TFLOPS	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
BF16 17.36 TFLOPS Tensor BF16 34.71 TFLOPS BF16 Tensor Sparse 69.43 TFLOPS	BF16 133.8 TFLOPS Tensor BF16 989.4 TFLOPS BF16 Tensor Sparse 1.98 PFLOPS
Tensor TF32 17.36 TFLOPS	Tensor TF32 494.7 TFLOPS
FP32 17.36 TFLOPS	FP32 66.91 TFLOPS
FP64 271.2 GFLOPS Tensor FP64 -	FP64 33.45 TFLOPS Tensor FP64 61.84 TFLOPS
Tensor INT4 277.7 TOPS	Tensor INT4 -
Tensor INT8 138.9 TOPS	Tensor INT8 1.98 POPS
Ray 26.18 TOPS	Ray -
Pixel Rate 54.2 GPixel/s	Pixel Rate 47.5 GPixel/s
Texture Rate 67.8 GTexel/s	Texture Rate 1.04 TTexel/s

Core Config

Shaders 5,120 Shaders	Shaders 16,896 Shaders
TMUs 160 TMUs	TMUs 528 TMUs
ROPs 128 ROPs	ROPs 24 ROPs
Tensor Cores 160 T-Cores	Tensor Cores 528 T-Cores
RT Cores 40 RT-Cores	RT Cores -
SMs 40 SMs	SMs 132 SMs

Clocks

Base Clock 885MHz	Base Clock 1.06GHz
Boost Clock 1.7GHz	Boost Clock 1.98GHz
Tensor Clock -	Tensor Clock 1.83GHz

Cache

L2 Cache 2MB shared

L2 Cache 51.2MB shared

Memory

64GB GDDR6	96GB HBM3
Memory Bus 128-bit	Memory Bus 5120-bit
Memory Speed 7.2GT/s	Memory Speed 5.2GT/s
Memory Bandwidth 463.4GB/s	Memory Bandwidth 3.35TB/s
ECC No	ECC No

Power & Thermals

TDP 250W

TDP 700W

Encoder

Encoder Model NVENC 7

-

Decoder

Decoder Model NVDEC 5

Decoder Model 7x NVDEC 5

Card

Form Factor PCIe Card	Form Factor SXM5
PCIe 2-Slots	PCIe -
Height 112 mm (4.41") Width 267 mm (10.51") Depth 40 mm (1.57")	-
Cooling Passive	Cooling Passive
Power Connectors 1x 8-Pin EPS	Power Connectors -

Chip

Manufacturer NVIDIA	Manufacturer NVIDIA
Chip Designer NVIDIA	Chip Designer NVIDIA
Architecture Ampere	Architecture Hopper
Family Server	Family Server
Branding	Branding
Codename NV177	Codename NV180
Chip Variant GA107-140-A1	Chip Variant -
Market Segment Server	Market Segment Server
Release Date Apr 12, 2021	Release Date Mar 22, 2022

Fabrication

Foundry Samsung	Foundry TSMC
Fabrication Node 8N	Fabrication Node 4N
Die Size 200mm²	Die Size 814mm²
Transistor Count 8.7 Billion	Transistor Count 80 Billion
Transistor Density 43.5 MTr/mm²	Transistor Density 98.28 MTr/mm²

Images

No images available

No images available

Competitors

NVIDIAA16 vs GPU AMDRadeon RX 9060 XT 16GB NVIDIAA16 vs GPU AMDRadeon RX 7600 XT NVIDIAA16 vs GPU AMDRadeon RX 9070 Show 12 more

NVIDIAH100 SXM vs GPU NVIDIAB100 NVIDIAH100 SXM vs GPU NVIDIAB200 NVIDIAH100 SXM vs GPU NVIDIAGB200 Show 12 more