AMD Radeon Instinct MI325X vs NVIDIA Quadro RTX 3000 X2 Mobile

Comparison AMD Radeon Instinct MI325X with 288 GB HBM3e and 19,456 cores vs NVIDIA Quadro RTX 3000 X2 Mobile with 6 GB GDDR6 and 1,920 cores.

Loading...

Performance Rating

AMD Radeon Instinct MI325X outperforms NVIDIA Quadro RTX 3000 X2 Mobile by 2,083.41% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

AMD Radeon Instinct MI325X

100.0

AMD Radeon Instinct MI325X

100.0
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

NVIDIA Quadro RTX 3000 X2 Mobile

4.6

NVIDIA Quadro RTX 3000 X2 Mobile

4.6

Expert Comparison

AMD Radeon Instinct MI325X обладает значительно большей пропускной способностью памяти (10.3 TB/s против 336.0 GB/s) и гораздо больше ядер (19456 против 1920). Это делает её предпочтительной для вычислительных задач и больших данных. В то время как NVIDIA Quadro RTX 3000 X2 Mobile более подходящ для мобильных решений и специализированных задач с меньшими требованиями, таких как CAD и 3D моделирование.

Contents:

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

Memory

Memory Size

🔥 +4,700% 288 ГБ
6 GB ×2 (12 ГБ)

Memory Type

HBM3e GDDR6

Memory Bandwidth

🔥 10.3 TB/s
336.0 GB/s ×2 (672 GB/s)

Memory Bus Width

8,192 бит 192 бит ×2 (384 бит)

ML Performance

FP16 (Half Precision)

🔥 +6,067% 653.7 TFLOPS
10.6 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +1,442% 81.72 TFLOPS
5.299 TFLOPS

FP64 (Double Precision)

🔥 +49,248% 81.72 TFLOPS
0.1656 TFLOPS

CUDA Cores

🔥 +913% 19,456
1,920 ×2 (3840)

RT Cores

No 30 ×2 (60)

Architecture & Compatibility

GPU Architecture

CDNA 3.0 Turing

SM (Streaming Multiprocessor)

No 30

PCIe Version

PCIe 5.0 x16 PCIe 3.0 x16

ML Software Support

CUDA Version

No 7.5

CUDA Toolkit (first supported)

v10

CUDA Toolkit status

Supported Supported

Clocks & Performance

Base Clock

🔥 +6% 1,000
945

Boost Clock

🔥 +52% 2,100
1,380

Memory Clock

🔥 +44% 2,525
1,750

Power Consumption

Recommended PSU

1400 W No

Power Connector

None None

TDP/TGP

1000 W
🔥 -84% 160 W

Rendering

Texture Units (TMU)

🔥 +913% 1,216
120 ×2 (240)

ROP

No 30 ×2 (60)

L2 Cache

🔥 +300% 16 MB
4 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 MXM Module

Release Date

Oct. 12, 2024 May 27, 2019

Display Outputs

No outputs
Portable Device Dependent

Renting is cheaper than buying