AMD Radeon Instinct MI325X vs PNY RTX 5090 Triple Fan

Comparison of AMD Radeon Instinct MI325X with 288 GB HBM3e and 19,456 cores vs PNY RTX 5090 Triple Fan with 32 GB GDDR7 and 21,760 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

AMD Radeon Instinct MI325X

AMD Radeon Instinct MI325X

RX 7900 XTX RX 7900 XTX
Instinct MI300X Instinct MI300X
MI250 MI250

PNY RTX 5090 Triple Fan

PNY RTX 5090 Triple Fan

Contents:

Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering Benchmarks Additional

Memory

Memory Size

288 GB 32 GB

Memory Type

HBM3e GDDR7

Memory Bandwidth

10.3 TB/s 1.79 TB/s

Memory Bus Width

8,192 бит 512 бит

ML Performance

FP16 (Half Precision)

653.7 TFLOPS 104.8 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

81.72 TFLOPS 104.8 TFLOPS

FP64 (Double Precision)

81.72 TFLOPS 1.637 TFLOPS

CUDA Cores

19,456 21,760

RT Cores

No 170

Architecture & Compatibility

GPU Architecture

CDNA 3.0 Blackwell 2.0

SM (Streaming Multiprocessor)

No 170

PCIe Version

PCIe 5.0 x16 PCIe 5.0 x16

ML Software Support

CUDA Version

No 12.0

Clocks & Performance

Base Clock

1,000 2,017

Boost Clock

2,100 2,407

Memory Clock

2,525 1,750

Power Consumption

TDP/TGP

1000 W
🔥 -42% 575 W

Recommended PSU

1400 W
🔥 -32% 950 W

Power Connector

None 1x 16-pin

Rendering

Texture Units (TMU)

1,216 680

ROP

No 170

L2 Cache

16 MB 96 MB

Benchmarks

MLPerf, llama2-70b-99.9 (Dummy)

3 596 tokens/s

MLPerf, llama2-70b-99.9 (fp8)

1 946 tokens/s

llama.cpp, llama-2-7b-Q4_0

22.4 tokens/s

MLPerf, mixtral-8x7b (fp8)

6 975 tokens/s

Additional

Slots

OAM Module Triple-slot

Release Date

Oct. 12, 2024 Jan. 30, 2025

Display Outputs

No outputs
1x HDMI 2.1b
3x DisplayPort 2.1b

Renting is cheaper than buying