AMD Instinct MI200 vs NVIDIA A100X Full Specs

Summary

6,656 Shaders 1.7GHz	6,912 Shaders 1.44GHz
64GB HBM2e1.64TB/s	80GB HBM2e2.04TB/s
Peak·FP32·FP16 22.63 TFLOPS	Peak·FP32·FP16 19.91 TFLOPS
Form Factor OAM Module	Form Factor PCIe Card
TDP 300W	TDP 300W
Power Connectors -	Power Connectors 1x 8-Pin EPS

Highlights

Peak OPS

Peak OPS FP32 FP16 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 Tensor INT4 Memory Bandwidth TDP

Instinct MI200181 TOPSINT4 Tensor

x1

A100X2.55 POPSINT4 Tensor Sparse

x14.07

Show Top Score

Theoretical Performance

Clock Speed1700MHz UC·Base·Boost·OC	Clock Speed1440MHz UC·Base·Boost·OC
Peak OPS 181 TOPSINT4 Tensor	Peak OPS 2.55 POPSINT4 Tensor Sparse
Tensor FP16-16 181 TFLOPS FP16-16 Tensor Sparse - Tensor FP16-32 181 TFLOPS FP16-32 Tensor Sparse -	Tensor FP16-16 318.5 TFLOPS FP16-16 Tensor Sparse 637 TFLOPS Tensor FP16-32 318.5 TFLOPS FP16-32 Tensor Sparse 637 TFLOPS
BF16 - Tensor BF16 181 TFLOPS BF16 Tensor Sparse -	BF16 39.81 TFLOPS Tensor BF16 318.5 TFLOPS BF16 Tensor Sparse 637 TFLOPS
Tensor TF32 -	Tensor TF32 159.3 TFLOPS
FP32 22.63 TFLOPS Tensor FP32 45.26 TFLOPS	FP32 19.91 TFLOPS Tensor FP32 -
FP64 22.63 TFLOPS Tensor FP64 45.26 TFLOPS	FP64 9.95 TFLOPS Tensor FP64 19.91 TFLOPS
Tensor INT4 181 TOPS	Tensor INT4 1.27 POPS
Tensor INT8 181 TOPS	Tensor INT8 637 TOPS
Pixel Rate -	Pixel Rate 230.4 GPixel/s
Texture Rate 707.2 GTexel/s	Texture Rate 622.1 GTexel/s

Core Config

Shaders 6,656 Shaders	Shaders 6,912 Shaders
TMUs 416 TMUs	TMUs 432 TMUs
ROPs -	ROPs 160 ROPs
Tensor Cores 416 T-Cores	Tensor Cores 432 T-Cores
CUs 104 CUs	SMs 108 SMs

Clocks

Base Clock 1GHz	Base Clock 795MHz
Boost Clock 1.7GHz	Boost Clock 1.44GHz

Cache

L2 Cache 8.2MB shared

L2 Cache 41MB shared

Memory

64GB HBM2e	80GB HBM2e
Memory Bus 4096-bit	Memory Bus 5120-bit
Memory Speed 3.2GT/s	Memory Speed 3.2GT/s
Memory Bandwidth 1.64TB/s	Memory Bandwidth 2.04TB/s
ECC Yes	ECC -

Power & Thermals

TDP 300W

TDP 300W

Encoder

Encoder Model VCN 2.6

-

Decoder

Decoder Model VCN 2.6

Decoder Model 5x NVDEC 4

Card

Form Factor OAM Module	Form Factor PCIe Card
PCIe -	PCIe 2-Slots
-	Height 111 mm (4.37") Width 267 mm (10.51") Depth 40 mm (1.57")
Cooling Passive	Cooling Passive
Power Connectors -	Power Connectors 1x 8-Pin EPS

Chip

Manufacturer AMD	Manufacturer NVIDIA
Chip Designer AMD	Chip Designer NVIDIA
Architecture CDNA 2	Architecture Ampere
Family Instinct	Family Server
Branding	Branding
Codename Aldebaran	Codename NV170
Chip Variant Aldebaran XL	Chip Variant -
Market Segment Server	Market Segment Server
Release Date Mar 22, 2022	Release Date Jun 28, 2021

Fabrication

Foundry TSMC	Foundry TSMC
Fabrication Node N6	Fabrication Node 7N
Die Size 724mm²	Die Size 826mm²
Transistor Count 28 Billion	Transistor Count 54.2 Billion
Transistor Density 38.67 MTr/mm²	Transistor Density 65.62 MTr/mm²

Images

No images available

No images available

Competitors

AMDInstinct MI200 vs GPU AMDInstinct MI300P AMDInstinct MI200 vs GPU AMDInstinct MI300X AMDInstinct MI200 vs GPU AMDInstinct MI350X Show 12 more

NVIDIAA100X vs GPU AMDInstinct MI300P NVIDIAA100X vs GPU NVIDIARTX PRO 6000 Blackwell Max-Q NVIDIAA100X vs GPU NVIDIAGeForce RTX 5070 Show 12 more