AMD Radeon Instinct MI250 vs KFA2 RTX 3090 HOF

Comparison of AMD Radeon Instinct MI250 with 128 GB HBM2e and 13,312 cores vs KFA2 RTX 3090 HOF with 24 GB GDDR6X and 10,496 cores.

Loading...

Performance Rating

AMD Radeon Instinct MI250 outperforms KFA2 RTX 3090 HOF by 147.76% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

AMD Radeon Instinct MI250

50.9

AMD Radeon Instinct MI250

50.9
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

KFA2 RTX 3090 HOF

20.5

KFA2 RTX 3090 HOF

20.5

Contents:

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

Memory

Memory Size

🔥 +433% 128 ГБ
24 ГБ

Memory Type

HBM2e GDDR6X

Memory Bandwidth

🔥 3.28 TB/s
936.2 GB/s

Memory Bus Width

8,192 бит 384 бит

ML Performance

FP16 (Half Precision)

🔥 +858% 362.1 TFLOPS
37.79 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +20% 45.26 TFLOPS
37.79 TFLOPS

FP64 (Double Precision)

🔥 +7,566% 45.26 TFLOPS
0.5903999999999999 TFLOPS

CUDA Cores

🔥 +27% 13,312
10,496

RT Cores

No
🔥 82

Architecture & Compatibility

GPU Architecture

CDNA 2.0 Ampere

SM (Streaming Multiprocessor)

No
🔥 82

PCIe Version

PCIe 4.0 x16 PCIe 4.0 x16

ML Software Support

CUDA Version

No 8.6

Clocks & Performance

Base Clock

1,000
🔥 +40% 1,395

Boost Clock

1,700
🔥 +6% 1,800

Memory Clock

🔥 +31% 1,600
1,219

Power Consumption

TDP/TGP

500 W
🔥 -30% 350 W

Recommended PSU

900 W
🔥 -17% 750 W

Power Connector

2x 8-pin 3x 8-pin

Rendering

Texture Units (TMU)

🔥 +154% 832
328

ROP

No
🔥 82

L2 Cache

16 MB
🔥 6 MB

Benchmarks

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

63.9 tokens/s

Additional

Slots

Dual-slot Triple-slot

Release Date

Nov. 8, 2021 Sept. 1, 2020

Display Outputs

No outputs
1x HDMI 2.1
3x DisplayPort 1.4a

Renting is cheaper than buying