AMD Radeon Instinct MI300X vs Intel Arc B570

Comparison AMD Radeon Instinct MI300X with 192 GB HBM3 and 19,456 cores vs Intel Arc B570 with 10 GB GDDR6 and 2,304 cores.

Loading...

Performance Rating

AMD Radeon Instinct MI300X outperforms Intel Arc B570 by 1,212.41% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

AMD Radeon Instinct MI300X

94.1

AMD Radeon Instinct MI300X

94.1
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

Intel Arc B570

7.2

Intel Arc B570

7.2

Expert Comparison

AMD Radeon Instinct MI300X значительно превосходит Intel Arc B570 по многим параметрам: количество ядер (19456 против 2304), объем памяти (192 ГБ HBM3 против 10 ГБ GDDR6), пропускную способность (10.3 ТБ/с против 380 ГБ/с) и индексу производительности (94.1 против 7.2). MI300X лучше подходит для высокопроизводительных вычислений и научных расчетов, тогда как Arc B570 более ориентирован на офисные задачи и легкие графические приложения.

Contents:

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

Memory

Memory Size

🔥 +1,820% 192 ГБ
10 GB

Memory Type

HBM3 GDDR6

Memory Bandwidth

🔥 10.3 TB/s
380.0 GB/s

Memory Bus Width

8,192 бит 160 бит

ML Performance

FP16 (Half Precision)

🔥 +2,737% 653.7 TFLOPS
23.04 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +609% 81.72 TFLOPS
11.52 TFLOPS

FP64 (Double Precision)

🔥 +11,250% 81.72 TFLOPS
0.72 TFLOPS

CUDA Cores

🔥 +744% 19,456
2,304

RT Cores

No
🔥 18

Architecture & Compatibility

GPU Architecture

CDNA 3.0 Xe2-HPG

SM (Streaming Multiprocessor)

No No

PCIe Version

PCIe 5.0 x16 PCIe 4.0 x8

ML Software Support

CUDA Version

No No

CUDA Toolkit status

Supported Supported

Clocks & Performance

Base Clock

1,000
🔥 +150% 2,500

Boost Clock

2,100
🔥 +19% 2,500

Memory Clock

🔥 +6% 2,525
2,375

Power Consumption

Recommended PSU

1150 W
🔥 -61% 450 W

Power Connector

None 1x 8-pin

TDP/TGP

750 W
🔥 -80% 150 W

Rendering

Texture Units (TMU)

🔥 +744% 1,216
144

ROP

No
🔥 18

L2 Cache

16 MB
🔥 +12% 18 MB

Benchmarks

MLPerf, llama2-70b-99.9 (UNSET)

1 983 tokens/s

MLPerf, llama2-70b-99.9 (fp16)

1 740 tokens/s

MLPerf, llama2-70b-99.9 (fp8)

1 057 tokens/s

MLPerf, llama3.1-405b (UNSET)

30.4 tokens/s

MLPerf, llama3.1-405b (fp16)

34.8 tokens/s

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

232.9 tokens/s

MLPerf, mixtral-8x7b (fp8)

5 975 tokens/s

Additional

Slots

OAM Module Dual-slot

Release Date

Dec. 6, 2023 Jan. 16, 2025

Display Outputs

No outputs
1x HDMI 2.1a
3x DisplayPort 2.1

Renting is cheaper than buying