GIGABYTE AORUS RTX 5090 INFINITY vs NVIDIA Switch GPU 16nm

Comparison of GIGABYTE AORUS RTX 5090 INFINITY with 32 GB GDDR7 and 21,760 cores vs NVIDIA Switch GPU 16nm with 4 GB DDR4 and 256 cores.

Loading...

Performance Rating

GIGABYTE AORUS RTX 5090 INFINITY outperforms NVIDIA Switch GPU 16nm by 8,975.0% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

GIGABYTE AORUS RTX 5090 INFINITY

47.2

GIGABYTE AORUS RTX 5090 INFINITY

47.2
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

NVIDIA Switch GPU 16nm

0.5

NVIDIA Switch GPU 16nm

0.5

Contents:

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

Memory

Memory Size

🔥 +700% 32 ГБ
4 GB

Memory Type

GDDR7 DDR4

Memory Bandwidth

🔥 1.79 TB/s
25.60 GB/s

Memory Bus Width

512 бит 64 бит

ML Performance

FP16 (Half Precision)

🔥 +15,007% 118.8 TFLOPS
0.7864 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +30,114% 118.8 TFLOPS
0.3932 TFLOPS

FP64 (Double Precision)

🔥 +14,989% 1.856 TFLOPS
0.0123 TFLOPS

CUDA Cores

🔥 +8,400% 21,760
256

RT Cores

🔥 170
No

Architecture & Compatibility

GPU Architecture

Blackwell 2.0 Maxwell 2.0

SM (Streaming Multiprocessor)

🔥 170
No

PCIe Version

PCIe 5.0 x16 No

ML Software Support

CUDA Version

🔥 12.0
5.3

Clocks & Performance

Base Clock

🔥 +425% 2,017
384

Boost Clock

🔥 +255% 2,730
768

Memory Clock

🔥 +9% 1,750
1,600

Power Consumption

Recommended PSU

950 W No

Power Connector

1x 16-pin No

TDP/TGP

575 W
🔥 -97% 15 W

Rendering

Texture Units (TMU)

🔥 +4,150% 680
16

ROP

🔥 170
No

L2 Cache

🔥 96 MB
No

Additional

Slots

Triple-slot No

Release Date

Jan. 30, 2025 Aug. 16, 2019

Display Outputs

1x HDMI 2.1b
3x DisplayPort 2.1b
1x USB Type-C

Renting is cheaper than buying