AMD Instinct MI300X vs NVIDIA B100 Full Specs

Summary

19,456 Shaders 2.1GHz	18,944 Shaders 1.6GHz
192GB HBM35.33TB/s	192GB HBM3e7.7TB/s
Peak·FP32·FP16 163.4 TFLOPS	Peak·FP32·FP16 60.62 TFLOPS
Form Factor OAM Module	Form Factor SXM6
TDP 560W	TDP 700W

Highlights

Peak OPS

Peak OPS FP32 FP16 Tensor FP4 FP4 Tensor Sparse 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

Instinct MI300X5.23 POPSINT4 Tensor Sparse

x1

B10014.01 PFLOPSFP4 Tensor Sparse

x2.68

Show Top Score

Theoretical Performance

Clock Speed2100MHz UC·Base·Boost·OC	Clock Speed1600MHz UC·Boost·OC
Peak OPS 5.23 POPSINT4 Tensor Sparse	Peak OPS 14.01 PFLOPSFP4 Tensor Sparse
-	Tensor FP4 7 PFLOPS FP4 Tensor Sparse 14.01 PFLOPS
Tensor FP8-16 2.62 PFLOPS FP8-16 Tensor Sparse 5.23 PFLOPS Tensor FP8-32 2.62 PFLOPS FP8-32 Tensor Sparse 5.23 PFLOPS	Tensor FP8-16 3.5 PFLOPS FP8-16 Tensor Sparse 7 PFLOPS Tensor FP8-32 3.5 PFLOPS FP8-32 Tensor Sparse 7 PFLOPS
Tensor FP16-16 1.31 PFLOPS FP16-16 Tensor Sparse 2.62 PFLOPS Tensor FP16-32 1.31 PFLOPS FP16-32 Tensor Sparse 2.62 PFLOPS	Tensor FP16-16 1.75 PFLOPS FP16-16 Tensor Sparse 3.5 PFLOPS Tensor FP16-32 1.75 PFLOPS FP16-32 Tensor Sparse 3.5 PFLOPS
BF16 - Tensor BF16 1.31 PFLOPS BF16 Tensor Sparse 2.62 PFLOPS	BF16 121.2 TFLOPS Tensor BF16 1.75 PFLOPS BF16 Tensor Sparse 3.5 PFLOPS
Tensor TF32 653.7 TFLOPS	Tensor TF32 875.4 TFLOPS
FP32 163.4 TFLOPS Tensor FP32 163.4 TFLOPS	FP32 60.62 TFLOPS Tensor FP32 -
FP64 81.72 TFLOPS Tensor FP64 163.4 TFLOPS	FP64 30.31 TFLOPS Tensor FP64 27.36 TFLOPS
Tensor INT4 2.62 POPS	Tensor INT4 -
Tensor INT8 2.62 POPS	Tensor INT8 3.5 POPS
Pixel Rate -	Pixel Rate 51.2 GPixel/s
Texture Rate 2.55 TTexel/s	Texture Rate 947.2 GTexel/s

Core Config

Shaders 19,456 Shaders	Shaders 18,944 Shaders
TMUs 1,216 TMUs	TMUs 592 TMUs
ROPs -	ROPs 32 ROPs
Tensor Cores 1,216 T-Cores	Tensor Cores 592 T-Cores
CUs 304 CUs	SMs 148 SMs

Clocks

Base Clock 1GHz	Base Clock -
Boost Clock 2.1GHz	Boost Clock 1.6GHz
Tensor Clock -	Tensor Clock 1.44GHz

Cache

L2 Cache 16.4MB shared	L2 Cache 65.5MB shared
L3 Cache 256MB shared	L3 Cache -

Memory

192GB HBM3	192GB HBM3e
Memory Bus 8192-bit	Memory Bus 8192-bit
Memory Speed 5.2GT/s	Memory Speed 7.5GT/s
Memory Bandwidth 5.33TB/s	Memory Bandwidth 7.7TB/s
ECC Yes	ECC No

Power & Thermals

TDP 560W

TDP 700W

Encoder

Encoder Model 2x VCN 2.6

-

Decoder

Decoder Model 2x VCN 2.6

Decoder Model 7x NVDEC 6

Card

Form Factor OAM Module	Form Factor SXM6
Cooling Passive	Cooling Passive

Chip

Manufacturer AMD	Manufacturer NVIDIA
Chip Designer AMD	Chip Designer NVIDIA
Architecture CDNA 3	Architecture Blackwell
Family Instinct	Family Server
Branding	Branding
Codename Aqua Vanjaram	Codename NV190
Chip Variant Aqua Vanjaram XTX	Chip Variant Umbriel
Market Segment Server	Market Segment Server
Release Date Dec 6, 2023	Release Date Mar 18, 2024

Fabrication

Foundry TSMC	Foundry TSMC
Other Foundries TSMCActive Interposer	Other Foundries -
Fabrication Node N5 N6Active Interposer	Fabrication Node 4NP -
Die Size 1480mm²	Die Size 1620mm²
Transistor Count 146 Billion	Transistor Count 208 Billion
Transistor Density 98.65 MTr/mm²	Transistor Density 128.4 MTr/mm²

Images

No images available

No images available

Competitors

AMDInstinct MI300X vs GPU AMDInstinct MI325X AMDInstinct MI300X vs GPU AMDInstinct MI350X AMDInstinct MI300X vs GPU AMDInstinct MI355X Show 12 more

NVIDIAB100 vs GPU NVIDIAB200 NVIDIAB100 vs GPU NVIDIAGB200 NVIDIAB100 vs GPU AMDInstinct MI325X Show 12 more