NVIDIA A100 vs AMD Instinct MI300P Full Specs

Summary

6,912 Shaders 1.41GHz	9,728 Shaders 2.1GHz
80GB HBM2e1.94TB/s	64GB HBM32.66TB/s
Peak·FP32·FP16 19.49 TFLOPS	Peak·FP32·FP16 81.72 TFLOPS
Form Factor PCIe Card	Form Factor OAM Module
TDP 250W	TDP 300W
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 Tensor FP32 FP64 Tensor FP64 BF16 Tensor BF16 BF16 Tensor Sparse Tensor TF32 Tensor INT8 Tensor INT4 Memory Bandwidth TDP

A1002.5 POPSINT4 Tensor Sparse

x1

Instinct MI300P2.62 POPSINT4 Tensor Sparse

x1.05

Show Top Score

Theoretical Performance

Clock Speed1410MHz UC·Base·Boost·OC	Clock Speed2100MHz UC·Base·Boost·OC
Peak OPS 2.5 POPSINT4 Tensor Sparse	Peak OPS 2.62 POPSINT4 Tensor Sparse
-	Tensor FP8-16 1.31 PFLOPS FP8-16 Tensor Sparse 2.62 PFLOPS Tensor FP8-32 1.31 PFLOPS FP8-32 Tensor Sparse 2.62 PFLOPS
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 FP16-16 653.7 TFLOPS FP16-16 Tensor Sparse 1.31 PFLOPS Tensor FP16-32 653.7 TFLOPS FP16-32 Tensor Sparse 1.31 PFLOPS
BF16 38.98 TFLOPS Tensor BF16 311.9 TFLOPS BF16 Tensor Sparse 623.7 TFLOPS	BF16 - Tensor BF16 653.7 TFLOPS BF16 Tensor Sparse 1.31 PFLOPS
Tensor TF32 155.9 TFLOPS	Tensor TF32 326.9 TFLOPS
FP32 19.49 TFLOPS Tensor FP32 -	FP32 81.72 TFLOPS Tensor FP32 81.72 TFLOPS
FP64 9.75 TFLOPS Tensor FP64 19.49 TFLOPS	FP64 40.86 TFLOPS Tensor FP64 81.72 TFLOPS
Tensor INT4 1.25 POPS	Tensor INT4 1.31 POPS
Tensor INT8 623.7 TOPS	Tensor INT8 1.31 POPS
Pixel Rate 225.6 GPixel/s	Pixel Rate -
Texture Rate 609.1 GTexel/s	Texture Rate 1.28 TTexel/s

Core Config

Shaders 6,912 Shaders	Shaders 9,728 Shaders
TMUs 432 TMUs	TMUs 608 TMUs
ROPs 160 ROPs	ROPs -
Tensor Cores 432 T-Cores	Tensor Cores 608 T-Cores
SMs 108 SMs	CUs 152 CUs

Clocks

Base Clock 765MHz	Base Clock 1GHz
Boost Clock 1.41GHz	Boost Clock 2.1GHz

Cache

L2 Cache 41MB shared	L2 Cache 16.4MB shared
L3 Cache -	L3 Cache 128MB shared

Memory

80GB HBM2e	64GB HBM3
Memory Bus 5120-bit	Memory Bus 4096-bit
Memory Speed 3GT/s	Memory Speed 5.2GT/s
Memory Bandwidth 1.94TB/s	Memory Bandwidth 2.66TB/s
ECC -	ECC Yes

Power & Thermals

TDP 250W

TDP 300W

Encoder

-	Encoder Model 2x VCN 2.6

Decoder

Decoder Model 5x NVDEC 4

Decoder Model 2x VCN 2.6

Card

Form Factor PCIe Card	Form Factor OAM Module
PCIe 2-Slots	PCIe -
Height 111 mm (4.37") 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 AMD
Chip Designer NVIDIA	Chip Designer AMD
Architecture Ampere	Architecture CDNA 3
Family Server	Family Instinct
Branding	Branding
Codename NV170	Codename Aqua Vanjaram
Chip Variant -	Chip Variant Aqua Vanjaram XL
Market Segment Server	Market Segment Server
Release Date Jun 28, 2021	Release Date Dec 6, 2023

Fabrication

Foundry TSMC	Foundry TSMC
Other Foundries -	Other Foundries TSMCActive Interposer
Fabrication Node 7N -	Fabrication Node N5 N6Active Interposer
Die Size 826mm²	Die Size 740mm²
Transistor Count 54.2 Billion	Transistor Count 73 Billion
Transistor Density 65.62 MTr/mm²	Transistor Density 98.65 MTr/mm²

Images

No images available

No images available

Competitors

NVIDIAA100 vs GPU NVIDIARTX PRO 6000 Blackwell Max-Q NVIDIAA100 vs GPU NVIDIAGeForce RTX 5070 NVIDIAA100 vs GPU AMDRadeon RX 7650 GRE Show 12 more

AMDInstinct MI300P vs GPU AMDInstinct MI300A AMDInstinct MI300P vs GPU NVIDIARTX PRO 5000 Blackwell AMDInstinct MI300P vs GPU NVIDIARTX 5880 Ada Show 12 more