AMD Instinct MI250 vs AMD Instinct MI300P Full Specs

Summary

13,312 Shaders 1.7GHz	9,728 Shaders 2.1GHz
128GB HBM2e3.28TB/s	64GB HBM32.66TB/s
Peak·FP32·FP16 45.26 TFLOPS	Peak·FP32·FP16 81.72 TFLOPS
Form Factor OAM Module	Form Factor OAM Module
TDP 500W	TDP 300W

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

Instinct MI250362.1 TOPSINT4 Tensor

x1

Instinct MI300P2.62 POPSINT4 Tensor Sparse

x7.22

Show Top Score

Theoretical Performance

Clock Speed1700MHz UC·Base·Boost·OC	Clock Speed2100MHz UC·Base·Boost·OC
Peak OPS 362.1 TOPSINT4 Tensor	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 FP16-16 362.1 TFLOPS FP16-16 Tensor Sparse - Tensor FP16-32 362.1 TFLOPS FP16-32 Tensor Sparse -	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
Tensor BF16 362.1 TFLOPS BF16 Tensor Sparse -	Tensor BF16 653.7 TFLOPS BF16 Tensor Sparse 1.31 PFLOPS
Tensor TF32 -	Tensor TF32 326.9 TFLOPS
FP32 45.26 TFLOPS Tensor FP32 90.52 TFLOPS	FP32 81.72 TFLOPS Tensor FP32 81.72 TFLOPS
FP64 45.26 TFLOPS Tensor FP64 90.52 TFLOPS	FP64 40.86 TFLOPS Tensor FP64 81.72 TFLOPS
Tensor INT4 362.1 TOPS	Tensor INT4 1.31 POPS
Tensor INT8 362.1 TOPS	Tensor INT8 1.31 POPS
Texture Rate 1.41 TTexel/s	Texture Rate 1.28 TTexel/s

Core Config

Shaders 13,312 Shaders	Shaders 9,728 Shaders
TMUs 832 TMUs	TMUs 608 TMUs
Tensor Cores 832 T-Cores	Tensor Cores 608 T-Cores
CUs 208 CUs	CUs 152 CUs

Clocks

Base Clock 1GHz	Base Clock 1GHz
Boost Clock 1.7GHz	Boost Clock 2.1GHz

Cache

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

Memory

128GB HBM2e	64GB HBM3
Memory Bus 8192-bit	Memory Bus 4096-bit
Memory Speed 3.2GT/s	Memory Speed 5.2GT/s
Memory Bandwidth 3.28TB/s	Memory Bandwidth 2.66TB/s
ECC Yes	ECC Yes

Power & Thermals

TDP 500W

TDP 300W

Encoder

Encoder Model 2x VCN 2.6

Encoder Model 2x VCN 2.6

Decoder

Decoder Model 2x VCN 2.6

Decoder Model 2x VCN 2.6

Card

Form Factor OAM Module	Form Factor OAM Module
Cooling Passive	Cooling Passive

Chip

Manufacturer AMD	Manufacturer AMD
Chip Designer AMD	Chip Designer AMD
Architecture CDNA 2	Architecture CDNA 3
Family Instinct	Family Instinct
Branding	Branding
Codename Aldebaran	Codename Aqua Vanjaram
Chip Variant Aldebaran XT	Chip Variant Aqua Vanjaram XL
Market Segment Server	Market Segment Server
Release Date Nov 8, 2021	Release Date Dec 6, 2023

Fabrication

Foundry TSMC	Foundry TSMC
Other Foundries -	Other Foundries TSMCActive Interposer
Fabrication Node N6 -	Fabrication Node N5 N6Active Interposer
Die Size 1448mm²	Die Size 740mm²
Transistor Count 56 Billion	Transistor Count 73 Billion
Transistor Density 38.67 MTr/mm²	Transistor Density 98.65 MTr/mm²

Images

No images available

No images available

Competitors

AMDInstinct MI250 vs GPU AMDInstinct MI250X AMDInstinct MI250 vs GPU AMDInstinct MI300A AMDInstinct MI250 vs GPU IntelGaudi 3 Show 12 more

AMDInstinct MI300P vs GPU AMDInstinct MI200 AMDInstinct MI300P vs GPU AMDInstinct MI210 AMDInstinct MI300P vs GPU AMDInstinct MI300A Show 12 more