AMD Radeon Instinct MI325X vs AMD Radeon RX Vega 56 Mobile

Comparison AMD Radeon Instinct MI325X with 288 GB HBM3e and 19,456 cores vs AMD Radeon RX Vega 56 Mobile with 8 GB HBM2 and 3,584 cores.

Loading...

Performance Rating

AMD Radeon Instinct MI325X outperforms AMD Radeon RX Vega 56 Mobile by 1,136.09% 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

AMD Radeon RX Vega 56 Mobile

8.1

AMD Radeon RX Vega 56 Mobile

8.1

Expert Comparison

AMD Radeon Instinct MI325X значительно превосходит AMD Radeon RX Vega 56 Mobile по многим параметрам. MI325X имеет гораздо больше ядер (19456 против 3584), больше памяти (288 ГБ против 8 ГБ) и большую пропускную способность (10.3 ТБ/с против 409.6 ГБ/с). Также MI325X рассчитан на более высокие нагрузки (1000 Вт против 120 Вт) и имеет значительно лучший GPI (100.0 против 8.1). MI325X подходит для вычислений и больших данных, тогда как RX Vega 56 Mobile лучше для мобильных устройств или игр.

Contents:

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

Memory

Memory Size

🔥 +3,500% 288 ГБ
8 GB

Memory Type

HBM3e HBM2

Memory Bandwidth

🔥 10.3 TB/s
409.6 GB/s

Memory Bus Width

8,192 бит 2,048 бит

ML Performance

FP16 (Half Precision)

🔥 +3,405% 653.7 TFLOPS
18.65 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +776% 81.72 TFLOPS
9.326 TFLOPS

FP64 (Double Precision)

🔥 +13,922% 81.72 TFLOPS
0.5828 TFLOPS

CUDA Cores

🔥 +443% 19,456
3,584

RT Cores

No No

Architecture & Compatibility

GPU Architecture

CDNA 3.0 GCN 5.0

SM (Streaming Multiprocessor)

No No

PCIe Version

PCIe 5.0 x16 PCIe 3.0 x16

ML Software Support

CUDA Version

No No

CUDA Toolkit status

Supported Supported

Clocks & Performance

Base Clock

🔥 1,000
1,138

Boost Clock

🔥 +61% 2,100
1,301

Memory Clock

🔥 +216% 2,525
800

Power Consumption

Recommended PSU

1400 W No

Power Connector

None None

TDP/TGP

1000 W
🔥 -88% 120 W

Rendering

Texture Units (TMU)

🔥 +443% 1,216
224

ROP

No No

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 June 1, 2018

Display Outputs

No outputs
1x HDMI 2.0b
3x DisplayPort 1.4a

Renting is cheaper than buying