AMD Instinct MI300A vs AMD Instinct MI355X Full Specs

Summary

14,592 Shaders 2.1GHz	16,384 Shaders 2.4GHz
128GB HBM35.33TB/s	288GB HBM3e8TB/s
Peak·FP32·FP16 122.6 TFLOPS	Peak·FP32·FP16 157.3 TFLOPS
Form Factor APU SH5 Socket	Form Factor OAM Module
TDP 560W	TDP 1400W

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

Instinct MI300A3.92 POPSINT4 Tensor Sparse

x1

Instinct MI355X20.13 PFLOPSFP4 Tensor Sparse

x5.13

Show Top Score

Theoretical Performance

Clock Speed2100MHz UC·Base·Boost·OC	Clock Speed2400MHz UC·Boost·OC
Peak OPS 3.92 POPSINT4 Tensor Sparse	Peak OPS 20.13 PFLOPSFP4 Tensor Sparse
-	Tensor FP4 10.07 PFLOPS FP4 Tensor Sparse 20.13 PFLOPS
Tensor FP8-16 1.96 PFLOPS FP8-16 Tensor Sparse 3.92 PFLOPS Tensor FP8-32 1.96 PFLOPS FP8-32 Tensor Sparse 3.92 PFLOPS	Tensor FP8-16 5.03 PFLOPS FP8-16 Tensor Sparse 10.07 PFLOPS Tensor FP8-32 5.03 PFLOPS FP8-32 Tensor Sparse 10.07 PFLOPS
Tensor FP16-16 980.6 TFLOPS FP16-16 Tensor Sparse 1.96 PFLOPS Tensor FP16-32 980.6 TFLOPS FP16-32 Tensor Sparse 1.96 PFLOPS	Tensor FP16-16 2.52 PFLOPS FP16-16 Tensor Sparse 5.03 PFLOPS Tensor FP16-32 2.52 PFLOPS FP16-32 Tensor Sparse 5.03 PFLOPS
Tensor BF16 980.6 TFLOPS BF16 Tensor Sparse 1.96 PFLOPS	Tensor BF16 2.52 PFLOPS BF16 Tensor Sparse 5.03 PFLOPS
Tensor TF32 490.3 TFLOPS	Tensor TF32 -
FP32 122.6 TFLOPS Tensor FP32 122.6 TFLOPS	FP32 157.3 TFLOPS Tensor FP32 157.3 TFLOPS
FP64 61.29 TFLOPS Tensor FP64 122.6 TFLOPS	FP64 78.64 TFLOPS Tensor FP64 78.64 TFLOPS
Tensor INT4 1.96 POPS	Tensor INT4 5.03 POPS
Tensor INT8 1.96 POPS	Tensor INT8 5.03 POPS
Texture Rate 1.92 TTexel/s	Texture Rate -

Core Config

Shaders 14,592 Shaders	Shaders 16,384 Shaders
TMUs 912 TMUs	TMUs -
Tensor Cores 912 T-Cores	Tensor Cores 1,024 T-Cores
CUs 228 CUs	CUs 256 CUs

Clocks

Base Clock 1GHz	Base Clock -
Boost Clock 2.1GHz	Boost Clock 2.4GHz

Cache

L2 Cache 16.4MB shared	L2 Cache 16.4MB shared
L3 Cache 192MB shared	L3 Cache 256MB shared

Memory

128GB HBM3	288GB HBM3e
Memory Bus 8192-bit	Memory Bus 8192-bit
Memory Speed 5.2GT/s	Memory Speed 7.8GT/s
Memory Bandwidth 5.33TB/s	Memory Bandwidth 8TB/s
ECC Yes	ECC Yes

Power & Thermals

TDP 560W

TDP 1400W

Encoder

Encoder Model 2x VCN 2.6

Encoder Model 4x VCN 4.0

Decoder

Decoder Model 2x VCN 2.6

Decoder Model 4x VCN 4.0

Card

Form Factor APU SH5 Socket	Form Factor OAM Module
Cooling Passive	Cooling Passive

Chip

Manufacturer AMD	Manufacturer AMD
Chip Designer AMD	Chip Designer AMD
Architecture CDNA 3	Architecture CDNA 4
Family Instinct	Family Instinct
Branding	Branding
Codename Aqua Vanjaram	Codename CDNA Next
Chip Variant Aqua Vanjaram XT	Chip Variant -
Market Segment Server	Market Segment Server
Release Date Dec 6, 2023	Release Date Jun 12, 2025

Fabrication

Foundry TSMC	Foundry TSMC
Other Foundries TSMCActive Interposer -	Other Foundries - TSMCIO
Fabrication Node N5 N6Active Interposer -	Fabrication Node N3P - N6IO
Die Size 1110mm²	Die Size 1448mm²
Transistor Count 109.5 Billion -	Transistor Count 139.2 Billion 46 BillionIO
Transistor Density 98.65 MTr/mm²	Transistor Density 96.13 MTr/mm²

Images

No images available

No images available

Competitors

AMDInstinct MI300A vs GPU AMDInstinct MI300X AMDInstinct MI300A vs GPU AMDInstinct MI250X AMDInstinct MI300A vs GPU AMDInstinct MI325X Show 12 more

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