NVIDIA A100 40GB vs NVIDIA A2 Full Specs

Summary

6,912 Shaders 1.41GHz	1,280 Shaders 1.77GHz
40GB HBM2e1.55TB/s	16GB GDDR6100.1GB/s
Peak·FP32·FP16 19.49 TFLOPS	Peak·FP32·FP16 4.53 TFLOPS
Form Factor PCIe Card	Form Factor PCIe Card
TDP 250W	TDP 60W
Power Connectors 1x 8-Pin EPS	Power Connectors -

Highlights

Peak OPS

Peak OPS FP32 FP16 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

A100 40GB2.5 POPSINT4 Tensor Sparse

x17.21

A2145 TOPSINT4 Tensor Sparse

x1

Show Top Score

Theoretical Performance

Clock Speed1410MHz UC·Base·Boost·OC	Clock Speed1770MHz UC·Base·Boost·OC
Peak OPS 2.5 POPSINT4 Tensor Sparse	Peak OPS 145 TOPSINT4 Tensor Sparse
Tensor FP16-16 311.9 TFLOPS FP16-16 Tensor Sparse 623.7 TFLOPS Tensor FP16-32 311.9 TFLOPS FP16-32 Tensor Sparse 623.7 TFLOPS	Tensor FP16-16 18.12 TFLOPS FP16-16 Tensor Sparse 36.25 TFLOPS Tensor FP16-32 9.06 TFLOPS FP16-32 Tensor Sparse 18.12 TFLOPS
BF16 38.98 TFLOPS Tensor BF16 311.9 TFLOPS BF16 Tensor Sparse 623.7 TFLOPS	BF16 4.53 TFLOPS Tensor BF16 9.06 TFLOPS BF16 Tensor Sparse 18.12 TFLOPS
Tensor TF32 155.9 TFLOPS	Tensor TF32 4.53 TFLOPS
FP32 19.49 TFLOPS	FP32 4.53 TFLOPS
FP64 9.75 TFLOPS Tensor FP64 19.49 TFLOPS	FP64 70.8 GFLOPS Tensor FP64 -
Tensor INT4 1.25 POPS	Tensor INT4 72.5 TOPS
Tensor INT8 623.7 TOPS	Tensor INT8 36.25 TOPS
Ray -	Ray 6.83 TOPS
Pixel Rate 225.6 GPixel/s	Pixel Rate 56.6 GPixel/s
Texture Rate 609.1 GTexel/s	Texture Rate 70.8 GTexel/s

Core Config

Shaders 6,912 Shaders	Shaders 1,280 Shaders
TMUs 432 TMUs	TMUs 40 TMUs
ROPs 160 ROPs	ROPs 32 ROPs
Tensor Cores 432 T-Cores	Tensor Cores 40 T-Cores
RT Cores -	RT Cores 10 RT-Cores
SMs 108 SMs	SMs 10 SMs

Clocks

Base Clock 765MHz	Base Clock 1.44GHz
Boost Clock 1.41GHz	Boost Clock 1.77GHz

Cache

L2 Cache 41MB shared

L2 Cache 2MB shared

Memory

40GB HBM2e	16GB GDDR6
Memory Bus 5120-bit	Memory Bus 128-bit
Memory Speed 2.4GT/s	Memory Speed 6.3GT/s
Memory Bandwidth 1.55TB/s	Memory Bandwidth 100.1GB/s

Power & Thermals

TDP 250W

TDP 60W

Encoder

-	Encoder Model NVENC 7

Decoder

Decoder Model 5x NVDEC 4

Decoder Model NVDEC 5

Card

Form Factor PCIe Card	Form Factor PCIe Card
PCIe 2-Slots	PCIe 1-Slots
Height 111 mm (4.37") Width 267 mm (10.51") Depth 40 mm (1.57")	Height 69 mm (2.72") Width 168 mm (6.61") Depth 20 mm (0.79")
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 Ampere
Family Server	Family Server
Branding	Branding
Codename NV170	Codename NV177
Chip Variant -	Chip Variant GA107-140-A1
Market Segment Server	Market Segment Server
Release Date Jun 28, 2021	Release Date Nov 10, 2021

Fabrication

Foundry TSMC	Foundry Samsung
Fabrication Node 7N	Fabrication Node 8N
Die Size 826mm²	Die Size 200mm²
Transistor Count 54.2 Billion	Transistor Count 8.7 Billion
Transistor Density 65.62 MTr/mm²	Transistor Density 43.5 MTr/mm²

Images

No images available

No images available

Competitors

NVIDIAA100 40GB vs GPU AMDInstinct MI300P NVIDIAA100 40GB vs GPU NVIDIARTX PRO 4500 Blackwell NVIDIAA100 40GB vs GPU NVIDIARTX PRO 5000 Blackwell Show 12 more

NVIDIAA2 vs GPU AMDRadeon PRO W7500 NVIDIAA2 vs GPU AMDRadeon RX 7400 NVIDIAA2 vs GPU AMDRadeon PRO W7400 Show 12 more