Compare XPUs

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

Select XPUs to Compare

Clear all (5)Jump to results

Filter by Vendor

Showing 72 XPUs • 5 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 A40	AMD MI350X	Biren Technology BR100	AMD MI300X	Baidu Kunlun II
Architecture	Ampere	CDNA 4	Biren GPU	CDNA 3	Kunlun Core
Form Factor	PCIe	OAM	—	OAM	—
VRAM	48 GB	288 GB	64 GB	192 GB	32 GB
Memory Bandwidth	696 GB/s	—	—	5,300 GB/s	—
TFLOPs (FP32)	37.4	—	—	163	—
TFLOPs (FP16)	150	—	1,000	1,307	256
TFLOPs	150	2,100	—	1,307	—
TFLOPs (FP8)	—	—	—	2,614	—
TDP	300 W	850 W	550 W	750 W	200 W
Launch Date	Oct 2020	Jun 2025	Aug 2022	Dec 2023	Aug 2021

Efficiency Metrics

Metric	A40	MI350X	BR100	MI300X	Kunlun II
TFLOPs per Watt (FP32-eq)	0.25	1.24	—	0.87	—
Memory Bandwidth per GB	14.5 GB/s	—	—	27.6 GB/s	—

Performance Equivalence

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

To match 1x NVIDIA A40

AMD MI350X

Compute (FP32-eq)

0.07x

MI350X is 14.00x faster

VRAM

0.17x

MI350X has 6.00x more

Biren Technology BR100

VRAM

0.75x

BR100 has 1.33x more

AMD MI300X

Compute (FP32-eq)

0.11x

MI300X is 8.71x faster

FP32 Compute

0.23x

MI300X is 4.36x faster

VRAM

0.25x

MI300X has 4.00x more

Memory Bandwidth

0.13x

MI300X has 7.61x more

Baidu Kunlun II

VRAM

1.50x

Need 1.50x Kunlun II

To match 1x AMD MI350X

NVIDIA A40

Compute (FP32-eq)

14.00x

Need 14.00x A40

VRAM

6.00x

Need 6.00x A40

Biren Technology BR100

VRAM

4.50x

Need 4.50x BR100

AMD MI300X

Compute (FP32-eq)

1.61x

Need 1.61x MI300X

VRAM

1.50x

Need 1.50x MI300X

Baidu Kunlun II

VRAM

9.00x

Need 9.00x Kunlun II

To match 1x Biren Technology BR100

NVIDIA A40

VRAM

1.33x

Need 1.33x A40

AMD MI350X

VRAM

0.22x

MI350X has 4.50x more

AMD MI300X

VRAM

0.33x

MI300X has 3.00x more

Baidu Kunlun II

VRAM

2.00x

Need 2.00x Kunlun II

To match 1x AMD MI300X

NVIDIA A40

Compute (FP32-eq)

8.71x

Need 8.71x A40

FP32 Compute

4.36x

Need 4.36x A40

VRAM

4.00x

Need 4.00x A40

Memory Bandwidth

7.61x

Need 7.61x A40

AMD MI350X

Compute (FP32-eq)

0.62x

MI350X is 1.61x faster

VRAM

0.67x

MI350X has 1.50x more

Biren Technology BR100

VRAM

3.00x

Need 3.00x BR100

Baidu Kunlun II

VRAM

6.00x

Need 6.00x Kunlun II

To match 1x Baidu Kunlun II

NVIDIA A40

VRAM

0.67x

A40 has 1.50x more

AMD MI350X

VRAM

0.11x

MI350X has 9.00x more

Biren Technology BR100

VRAM

0.50x

BR100 has 2.00x more

AMD MI300X

VRAM

0.17x

MI300X has 6.00x more

Pricing

Price Type	A40	MI350X	BR100	MI300X	Kunlun II
CAPEX (Street Price)	—	—	—	$35,000	—
OPEX (per hour)	—	—	—	$10.40/hr	—
Price per TFLOPs (FP32-eq)	—	—	—	$54	—