AMD Radeon Instinct MI325X vs AMD Radeon R7 M260

Comparison AMD Radeon Instinct MI325X with 288 GB HBM3e and 19,456 cores vs AMD Radeon R7 M260 with 2 GB DDR3 and 384 cores.

Loading...

Performance Rating

AMD Radeon Instinct MI325X outperforms AMD Radeon R7 M260 by 16,566.67% 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 R7 M260

0.6

AMD Radeon R7 M260

0.6

Expert Comparison

AMD Radeon Instinct MI325X значительно превосходит AMD Radeon R7 M260 по всем параметрам. MI325X имеет гораздо больше ядер (19456 против 384), больше памяти (288 ГБ HBM3e против 2 ГБ DDR3) и большую пропускную способность (10.3 TB/s против 14.40 GB/s). Кроме того, MI325X предназначен для современных задач, таких как машинное обучение и вычислительная наука, тогда как R7 M260 более подходящ для базовых графических задач и игр на старых платформах.

Contents:

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

Memory

Memory Size

🔥 +14,300% 288 ГБ
2 GB

Memory Type

HBM3e DDR3

Memory Bandwidth

🔥 10.3 TB/s
14.40 GB/s

Memory Bus Width

8,192 бит 64 бит

ML Performance

FP16 (Half Precision)

🔥 +86,759% 653.7 TFLOPS
0.7526 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +10,758% 81.72 TFLOPS
0.7526 TFLOPS

FP64 (Double Precision)

🔥 +173,772% 81.72 TFLOPS
0.047 TFLOPS

CUDA Cores

🔥 +4,967% 19,456
384

RT Cores

No No

Architecture & Compatibility

GPU Architecture

CDNA 3.0 GCN 3.0

SM (Streaming Multiprocessor)

No No

PCIe Version

PCIe 5.0 x16 PCIe 3.0 x8

ML Software Support

CUDA Version

No No

CUDA Toolkit status

Supported Supported

Clocks & Performance

Base Clock

🔥 +6% 1,000
940

Boost Clock

🔥 +114% 2,100
980

Memory Clock

🔥 +181% 2,525
900

Power Consumption

Recommended PSU

1400 W No

Power Connector

None No

TDP/TGP

1000 W No

Rendering

Texture Units (TMU)

🔥 +4,967% 1,216
24

ROP

No No

L2 Cache

🔥 16 MB
128 KB

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 No

Release Date

Oct. 12, 2024 June 11, 2014

Display Outputs

No outputs
No

Renting is cheaper than buying