Compare XPUs

Select up to 5 XPUs to compare side-by-side

Select XPUs to Compare

Clear all (3)Jump to results

Filter by Vendor

Showing 128 XPUs • 3 selected

Alibaba

Hanguang 800

AMD

MI100

23.1 TFLOPs

AMD

MI210

181 TFLOPs

AMD

MI250X

383 TFLOPs

AMD

MI300X

1,307 TFLOPs

AMD

MI325X

1,400 TFLOPs

AMD

MI350X

2,100 TFLOPs

AMD

MI355X

5,300 TFLOPs

AMD

Radeon Pro V520

23.04 TFLOPs

AWS

Inferentia2

190 TFLOPs

AWS

Trainium

190 TFLOPs

AWS

Trainium2

680 TFLOPs

Baidu

Kunlun II

Biren Technology

BR100

Cambricon

MLU370

256 TFLOPs

Cerebras

WSE-3

Enflame Technology

CloudBlazer T20

FuriosaAI

RNGD (Renegade)

256 TFLOPs

FuriosaAI

Warboy

Google

TPU v4

275 TFLOPs

Google

TPU v5e

197 TFLOPs

Google

TPU v5p

459 TFLOPs

Graphcore

Bow IPU

Graphcore

IPU-M2000

Groq

LPU Inference Engine

Huawei

Ascend 910B

Iluvatar CoreX

BI-V150

300 TFLOPs

Intel

Data Center GPU Max 1100

177 TFLOPs

Intel

Data Center GPU Max 1550

419 TFLOPs

Intel Habana

Gaudi 2

432 TFLOPs

Intel Habana

Gaudi 3

1,835 TFLOPs

Multi-Metric Comparison

Relative performance across 5 key metrics (normalized to 100 = best in comparison)

Compute Performance (BF16)

Memory Capacity

Power Consumption

Power Efficiency

Specifications

Specification	NVIDIA GeForce GTX 1660	NVIDIA A100 SXM	AMD Radeon Pro V520
Architecture	Turing	Ampere	RDNA 2
Form Factor	PCIe	SXM	PCIe
VRAM	6 GB	80 GB	32 GB
Memory Bandwidth	192 GB/s	2,039 GB/s	448 GB/s
TFLOPs (FP32)	5.027	19.5	10.1
TFLOPs (FP16)	—	312	—
TFLOPs	5.027	312	23.04
TFLOPs (FP8)	—	—	—
TDP	120 W	400 W	225 W
Launch Date	Mar 2019	May 2020	Jun 2021

Efficiency Metrics

Metric	GeForce GTX 1660	A100 SXM	Radeon Pro V520
TFLOPs per Watt (FP32-eq)	0.04	0.39	0.05
Memory Bandwidth per GB	32.0 GB/s	25.5 GB/s	14.0 GB/s

Performance Equivalence

How many units of each GPU are needed to match the performance of the others?

To match 1x NVIDIA GeForce GTX 1660

NVIDIA A100 SXM

Compute (FP32-eq)

0.03x

A100 SXM is 31.03x faster

FP32 Compute

0.26x

A100 SXM is 3.88x faster

VRAM

0.07x

A100 SXM has 13.33x more

Memory Bandwidth

0.09x

A100 SXM has 10.62x more

AMD Radeon Pro V520

Compute (FP32-eq)

0.44x

Radeon Pro V520 is 2.29x faster

FP32 Compute

0.50x

Radeon Pro V520 is 2.01x faster

VRAM

0.19x

Radeon Pro V520 has 5.33x more

Memory Bandwidth

0.43x

Radeon Pro V520 has 2.33x more

To match 1x NVIDIA A100 SXM

NVIDIA GeForce GTX 1660

Compute (FP32-eq)

31.03x

Need 31.03x GeForce GTX 1660

FP32 Compute

3.88x

Need 3.88x GeForce GTX 1660

VRAM

13.33x

Need 13.33x GeForce GTX 1660

Memory Bandwidth

10.62x

Need 10.62x GeForce GTX 1660

AMD Radeon Pro V520

Compute (FP32-eq)

13.54x

Need 13.54x Radeon Pro V520

FP32 Compute

1.93x

Need 1.93x Radeon Pro V520

VRAM

2.50x

Need 2.50x Radeon Pro V520

Memory Bandwidth

4.55x

Need 4.55x Radeon Pro V520

To match 1x AMD Radeon Pro V520

NVIDIA GeForce GTX 1660

Compute (FP32-eq)

2.29x

Need 2.29x GeForce GTX 1660

FP32 Compute

2.01x

Need 2.01x GeForce GTX 1660

VRAM

5.33x

Need 5.33x GeForce GTX 1660

Memory Bandwidth

2.33x

Need 2.33x GeForce GTX 1660

NVIDIA A100 SXM

Compute (FP32-eq)

0.07x

A100 SXM is 13.54x faster

FP32 Compute

0.52x

A100 SXM is 1.93x faster

VRAM

0.40x

A100 SXM has 2.50x more

Memory Bandwidth

0.22x

A100 SXM has 4.55x more

Pricing

Price Type	GeForce GTX 1660	A100 SXM	Radeon Pro V520
CAPEX (Street Price)	—	$15,000	—
OPEX (per hour)	$0.04/hr	$4.05/hr	$0.38/hr
Price per TFLOPs (FP32-eq)	—	$96	—