AMD Instinct MI350X vs AMD Instinct MI325X Full Specs

Summary

16,384 Shaders 2.2GHz	19,456 Shaders 2.1GHz
288GB HBM3e8TB/s	288GB HBM3e6.01TB/s
Peak·FP32·FP16 144.2 TFLOPS	Peak·FP32·FP16 163.4 TFLOPS
Form Factor OAM Module	Form Factor OAM Module
TDP 1000W	TDP 560W

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 MI350X18.45 PFLOPSFP4 Tensor Sparse

x3.53

Instinct MI325X5.23 POPSINT4 Tensor Sparse

x1

Show Top Score

Theoretical Performance

Clock Speed2200MHz UC·Boost·OC	Clock Speed2100MHz UC·Base·Boost·OC
Peak OPS 18.45 PFLOPSFP4 Tensor Sparse	Peak OPS 5.23 POPSINT4 Tensor Sparse
Tensor FP4 9.23 PFLOPS FP4 Tensor Sparse 18.45 PFLOPS	-
Tensor FP8-16 4.61 PFLOPS FP8-16 Tensor Sparse 9.23 PFLOPS Tensor FP8-32 4.61 PFLOPS FP8-32 Tensor Sparse 9.23 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 FP16-16 2.31 PFLOPS FP16-16 Tensor Sparse 4.61 PFLOPS Tensor FP16-32 2.31 PFLOPS FP16-32 Tensor Sparse 4.61 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 BF16 2.31 PFLOPS BF16 Tensor Sparse 4.61 PFLOPS	Tensor BF16 1.31 PFLOPS BF16 Tensor Sparse 2.62 PFLOPS
Tensor TF32 -	Tensor TF32 653.7 TFLOPS
FP32 144.2 TFLOPS Tensor FP32 144.2 TFLOPS	FP32 163.4 TFLOPS Tensor FP32 163.4 TFLOPS
FP64 72.09 TFLOPS Tensor FP64 72.09 TFLOPS	FP64 81.72 TFLOPS Tensor FP64 163.4 TFLOPS
Tensor INT4 4.61 POPS	Tensor INT4 2.62 POPS
Tensor INT8 4.61 POPS	Tensor INT8 2.62 POPS
Texture Rate -	Texture Rate 2.55 TTexel/s

Core Config

Shaders 16,384 Shaders	Shaders 19,456 Shaders
TMUs -	TMUs 1,216 TMUs
Tensor Cores 1,024 T-Cores	Tensor Cores 1,216 T-Cores
CUs 256 CUs	CUs 304 CUs

Clocks

Base Clock -	Base Clock 1GHz
Boost Clock 2.2GHz	Boost Clock 2.1GHz

Cache

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

Memory

288GB HBM3e	288GB HBM3e
Memory Bus 8192-bit	Memory Bus 8192-bit
Memory Speed 7.8GT/s	Memory Speed 5.9GT/s
Memory Bandwidth 8TB/s	Memory Bandwidth 6.01TB/s
ECC Yes	ECC Yes

Power & Thermals

TDP 1000W

TDP 560W

Encoder

Encoder Model 4x VCN 4.0

Encoder Model 2x VCN 2.6

Decoder

Decoder Model 4x VCN 4.0

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 4	Architecture CDNA 3
Family Instinct	Family Instinct
Branding	Branding
Codename CDNA Next	Codename Aqua Vanjaram
Chip Variant -	Chip Variant Aqua Vanjaram XTX
Market Segment Server	Market Segment Server
Release Date Jun 12, 2025	Release Date Jun 2, 2024

Fabrication

Foundry TSMC	Foundry TSMC
Other Foundries - TSMCIO	Other Foundries TSMCActive Interposer -
Fabrication Node N3P - N6IO	Fabrication Node N5 N6Active Interposer -
Die Size 1448mm²	Die Size 1480mm²
Transistor Count 139.2 Billion 46 BillionIO	Transistor Count 146 Billion -
Transistor Density 96.13 MTr/mm²	Transistor Density 98.65 MTr/mm²

Images

No images available

No images available

Competitors

AMDInstinct MI350X vs GPU AMDInstinct MI355X AMDInstinct MI350X vs GPU AMDInstinct MI300X AMDInstinct MI350X vs GPU AMDInstinct MI300A Show 12 more

AMDInstinct MI325X vs GPU AMDInstinct MI300X AMDInstinct MI325X vs GPU AMDInstinct MI355X AMDInstinct MI325X vs GPU AMDInstinct MI300A Show 12 more