Compare XPUs

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

Select XPUs to Compare

Clear all (4)Jump to results

Filter by Vendor

Showing 128 XPUs • 4 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	AMD MI325X	NVIDIA GeForce RTX 4060 Ti 16GB	NVIDIA GeForce GTX 1050 Ti	NVIDIA RTX A4000
Architecture	CDNA 3.5	Ada Lovelace	Pascal	Ampere
Form Factor	OAM	PCIe	PCIe	PCIe
VRAM	256 GB	16 GB	4 GB	16 GB
Memory Bandwidth	6,000 GB/s	288 GB/s	112 GB/s	448 GB/s
TFLOPs (FP32)	207	—	2.138	19.2
TFLOPs (FP16)	1,400	—	—	—
TFLOPs	1,400	44	2.138	38.7
TFLOPs (FP8)	2,800	—	—	—
TDP	800 W	165 W	75 W	140 W
Launch Date	Oct 2024	Jul 2024	Oct 2016	Apr 2021

Efficiency Metrics

Metric	MI325X	GeForce RTX 4060 Ti 16GB	GeForce GTX 1050 Ti	RTX A4000
TFLOPs per Watt (FP32-eq)	0.88	0.13	0.03	0.14
Memory Bandwidth per GB	23.4 GB/s	18.0 GB/s	28.0 GB/s	28.0 GB/s

Performance Equivalence

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

To match 1x AMD MI325X

NVIDIA GeForce RTX 4060 Ti 16GB

Compute (FP32-eq)

31.82x

Need 31.82x GeForce RTX 4060 Ti 16GB

VRAM

16.00x

Need 16.00x GeForce RTX 4060 Ti 16GB

Memory Bandwidth

20.83x

Need 20.83x GeForce RTX 4060 Ti 16GB

NVIDIA GeForce GTX 1050 Ti

Compute (FP32-eq)

327.41x

Need 327.41x GeForce GTX 1050 Ti

FP32 Compute

96.82x

Need 96.82x GeForce GTX 1050 Ti

VRAM

64.00x

Need 64.00x GeForce GTX 1050 Ti

Memory Bandwidth

53.57x

Need 53.57x GeForce GTX 1050 Ti

NVIDIA RTX A4000

Compute (FP32-eq)

36.18x

Need 36.18x RTX A4000

FP32 Compute

10.78x

Need 10.78x RTX A4000

VRAM

16.00x

Need 16.00x RTX A4000

Memory Bandwidth

13.39x

Need 13.39x RTX A4000

To match 1x NVIDIA GeForce RTX 4060 Ti 16GB

AMD MI325X

Compute (FP32-eq)

0.03x

MI325X is 31.82x faster

VRAM

0.06x

MI325X has 16.00x more

Memory Bandwidth

0.05x

MI325X has 20.83x more

NVIDIA GeForce GTX 1050 Ti

Compute (FP32-eq)

10.29x

Need 10.29x GeForce GTX 1050 Ti

VRAM

4.00x

Need 4.00x GeForce GTX 1050 Ti

Memory Bandwidth

2.57x

Need 2.57x GeForce GTX 1050 Ti

NVIDIA RTX A4000

Compute (FP32-eq)

1.14x

Need 1.14x RTX A4000

VRAM

1.00x

RTX A4000 has 1.00x more

Memory Bandwidth

0.64x

RTX A4000 has 1.56x more

To match 1x NVIDIA GeForce GTX 1050 Ti

AMD MI325X

Compute (FP32-eq)

0.00x

MI325X is 327.41x faster

FP32 Compute

0.01x

MI325X is 96.82x faster

VRAM

0.02x

MI325X has 64.00x more

Memory Bandwidth

0.02x

MI325X has 53.57x more

NVIDIA GeForce RTX 4060 Ti 16GB

Compute (FP32-eq)

0.10x

GeForce RTX 4060 Ti 16GB is 10.29x faster

VRAM

0.25x

GeForce RTX 4060 Ti 16GB has 4.00x more

Memory Bandwidth

0.39x

GeForce RTX 4060 Ti 16GB has 2.57x more

NVIDIA RTX A4000

Compute (FP32-eq)

0.11x

RTX A4000 is 9.05x faster

FP32 Compute

0.11x

RTX A4000 is 8.98x faster

VRAM

0.25x

RTX A4000 has 4.00x more

Memory Bandwidth

0.25x

RTX A4000 has 4.00x more

To match 1x NVIDIA RTX A4000

AMD MI325X

Compute (FP32-eq)

0.03x

MI325X is 36.18x faster

FP32 Compute

0.09x

MI325X is 10.78x faster

VRAM

0.06x

MI325X has 16.00x more

Memory Bandwidth

0.07x

MI325X has 13.39x more

NVIDIA GeForce RTX 4060 Ti 16GB

Compute (FP32-eq)

0.88x

GeForce RTX 4060 Ti 16GB is 1.14x faster

VRAM

1.00x

GeForce RTX 4060 Ti 16GB has 1.00x more

Memory Bandwidth

1.56x

Need 1.56x GeForce RTX 4060 Ti 16GB

NVIDIA GeForce GTX 1050 Ti

Compute (FP32-eq)

9.05x

Need 9.05x GeForce GTX 1050 Ti

FP32 Compute

8.98x

Need 8.98x GeForce GTX 1050 Ti

VRAM

4.00x

Need 4.00x GeForce GTX 1050 Ti

Memory Bandwidth

4.00x

Need 4.00x GeForce GTX 1050 Ti

Pricing

Price Type	MI325X	GeForce RTX 4060 Ti 16GB	GeForce GTX 1050 Ti	RTX A4000
CAPEX (Street Price)	$45,000	—	—	—
OPEX (per hour)	$36.92/hr	—	$0.04/hr	$0.11/hr
Price per TFLOPs (FP32-eq)	$64	—	—	—