AMD Radeon Instinct MI355X vs NVIDIA Quadro Plex 1000 Model IV

Comparison of AMD Radeon Instinct MI355X with 288 GB HBM3e and 16,384 cores vs NVIDIA Quadro Plex 1000 Model IV with 1,536 GB GDDR3 and 128 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

AMD Radeon Instinct MI355X

AMD Radeon Instinct MI355X

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

NVIDIA Quadro Plex 1000 Model IV

NVIDIA Quadro Plex 1000 Model IV

Contents:

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

Memory

Memory Size

288 GB 1,536 MB ×2 (3072 ГБ)

Memory Type

HBM3e GDDR3

Memory Bandwidth

8.19 TB/s 76.80 GB/s ×2 (153.6 GB/s)

Memory Bus Width

8,192 бит 384 бит ×2 (768 бит)

ML Performance

FP16 (Half Precision)

629.1 TFLOPS No

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

78.64 TFLOPS 0.3456 TFLOPS

FP64 (Double Precision)

78.64 TFLOPS No

CUDA Cores

16,384 128 ×2 (256)

RT Cores

No No

Architecture & Compatibility

GPU Architecture

CDNA 4.0 Tesla

SM (Streaming Multiprocessor)

No 16

PCIe Version

PCIe 5.0 x16 PCIe 1.0 x16

ML Software Support

CUDA Version

No 1.0

Clocks & Performance

Base Clock

1,000 No

Boost Clock

2,400 No

Memory Clock

2,000 800

Power Consumption

TDP/TGP

1400 W
🔥 -77% 320 W

Recommended PSU

1800 W
🔥 -61% 700 W

Power Connector

None No

Rendering

Texture Units (TMU)

1,024 32 ×2 (64)

ROP

No No

L2 Cache

32 MB 96 KB

Benchmarks

MLPerf, llama2-70b-99.9 (mxfp4)

10 129 tokens/s

MLPerf, llama3.1-405b (mxfp4)

252.4 tokens/s

Additional

Slots

OAM Module No

Release Date

March 12, 2025 July 25, 2008

Display Outputs

No outputs
No outputs

Renting is cheaper than buying