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 72 XPUs • 4 selected

Alibaba

Hanguang 800

AMD

MI100

23.1 TFLOPs

AMD

MI210

181 TFLOPs

AMD

MI250X

383 TFLOPs

AMD

MI300A

980.6 TFLOPs

AMD

MI300X

1,307 TFLOPs

AMD

MI325X

1,400 TFLOPs

AMD

MI350X

2,100 TFLOPs

AMD

Radeon PRO W7900

122 TFLOPs

AMD

Radeon RX 7900 XT

104 TFLOPs

AMD

Radeon RX 7900 XTX

122 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

Etched

Sohu

10,000 TFLOPs

FuriosaAI

Warboy

Google

TPU v4

275 TFLOPs

Google

TPU v5e

197 TFLOPs

Google

TPU v5p

459 TFLOPs

Google

TPU v6e (Trillium)

918 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 A100 SXM	AMD MI350X	NVIDIA GeForce RTX 4090	Baidu Kunlun II
Architecture	Ampere	CDNA 4	Ada Lovelace	Kunlun Core
Form Factor	SXM	OAM	PCIe	—
VRAM	80 GB	288 GB	24 GB	32 GB
Memory Bandwidth	2,039 GB/s	—	1,008 GB/s	—
TFLOPs (FP32)	19.5	—	82.6	—
TFLOPs (FP16)	312	—	165.2	256
TFLOPs	312	2,100	165.2	—
TFLOPs (FP8)	—	—	—	—
TDP	400 W	850 W	450 W	200 W
Launch Date	May 2020	Jun 2025	Oct 2022	Aug 2021

Efficiency Metrics

Metric	A100 SXM	MI350X	GeForce RTX 4090	Kunlun II
TFLOPs per Watt (FP32-eq)	0.39	1.24	0.18	—
Memory Bandwidth per GB	25.5 GB/s	—	42.0 GB/s	—

Performance Equivalence

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

To match 1x NVIDIA A100 SXM

AMD MI350X

Compute (FP32-eq)

0.15x

MI350X is 6.73x faster

VRAM

0.28x

MI350X has 3.60x more

NVIDIA GeForce RTX 4090

Compute (FP32-eq)

1.89x

Need 1.89x GeForce RTX 4090

FP32 Compute

0.24x

GeForce RTX 4090 is 4.24x faster

VRAM

3.33x

Need 3.33x GeForce RTX 4090

Memory Bandwidth

2.02x

Need 2.02x GeForce RTX 4090

Baidu Kunlun II

VRAM

2.50x

Need 2.50x Kunlun II

To match 1x AMD MI350X

NVIDIA A100 SXM

Compute (FP32-eq)

6.73x

Need 6.73x A100 SXM

VRAM

3.60x

Need 3.60x A100 SXM

NVIDIA GeForce RTX 4090

Compute (FP32-eq)

12.71x

Need 12.71x GeForce RTX 4090

VRAM

12.00x

Need 12.00x GeForce RTX 4090

Baidu Kunlun II

VRAM

9.00x

Need 9.00x Kunlun II

To match 1x NVIDIA GeForce RTX 4090

NVIDIA A100 SXM

Compute (FP32-eq)

0.53x

A100 SXM is 1.89x faster

FP32 Compute

4.24x

Need 4.24x A100 SXM

VRAM

0.30x

A100 SXM has 3.33x more

Memory Bandwidth

0.49x

A100 SXM has 2.02x more

AMD MI350X

Compute (FP32-eq)

0.08x

MI350X is 12.71x faster

VRAM

0.08x

MI350X has 12.00x more

Baidu Kunlun II

VRAM

0.75x

Kunlun II has 1.33x more

To match 1x Baidu Kunlun II

NVIDIA A100 SXM

VRAM

0.40x

A100 SXM has 2.50x more

AMD MI350X

VRAM

0.11x

MI350X has 9.00x more

NVIDIA GeForce RTX 4090

VRAM

1.33x

Need 1.33x GeForce RTX 4090

Pricing

Price Type	A100 SXM	MI350X	GeForce RTX 4090	Kunlun II
CAPEX (Street Price)	$15,000	—	—	—
OPEX (per hour)	$4.05/hr	—	—	—
Price per TFLOPs (FP32-eq)	$96	—	—	—