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	Cerebras WSE-3	SambaNova SN40L	Etched Sohu
Architecture	Ampere	CDNA 4	Wafer Scale Engine	Reconfigurable Dataflow Unit (RDU)	Transformer ASIC
Form Factor	PCIe	OAM	—	—	PCIe
VRAM	48 GB	288 GB	44 GB	640 GB	144 GB
Memory Bandwidth	696 GB/s	—	—	—	4,800 GB/s
TFLOPs (FP32)	37.4	—	—	—	10,000
TFLOPs (FP16)	150	—	—	—	—
TFLOPs	150	2,100	—	—	10,000
TFLOPs (FP8)	—	—	—	—	—
TDP	300 W	850 W	23000 W	700 W	700 W
Launch Date	Oct 2020	Jun 2025	Mar 2024	Jan 2024	Jun 2024

Efficiency Metrics

Metric	A40	MI350X	WSE-3	SN40L	Sohu
TFLOPs per Watt (FP32-eq)	0.25	1.24	—	—	14.29
Memory Bandwidth per GB	14.5 GB/s	—	—	—	33.3 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

Cerebras WSE-3

VRAM

1.09x

Need 1.09x WSE-3

SambaNova SN40L

VRAM

0.07x

SN40L has 13.33x more

Etched Sohu

Compute (FP32-eq)

0.01x

Sohu is 133.33x faster

FP32 Compute

0.00x

Sohu is 267.38x faster

VRAM

0.33x

Sohu has 3.00x more

Memory Bandwidth

0.14x

Sohu has 6.90x more

To match 1x AMD MI350X

NVIDIA A40

Compute (FP32-eq)

14.00x

Need 14.00x A40

VRAM

6.00x

Need 6.00x A40

Cerebras WSE-3

VRAM

6.55x

Need 6.55x WSE-3

SambaNova SN40L

VRAM

0.45x

SN40L has 2.22x more

Etched Sohu

Compute (FP32-eq)

0.10x

Sohu is 9.52x faster

VRAM

2.00x

Need 2.00x Sohu

To match 1x Cerebras WSE-3

NVIDIA A40

VRAM

0.92x

A40 has 1.09x more

AMD MI350X

VRAM

0.15x

MI350X has 6.55x more

SambaNova SN40L

VRAM

0.07x

SN40L has 14.55x more

Etched Sohu

VRAM

0.31x

Sohu has 3.27x more

To match 1x SambaNova SN40L

NVIDIA A40

VRAM

13.33x

Need 13.33x A40

AMD MI350X

VRAM

2.22x

Need 2.22x MI350X

Cerebras WSE-3

VRAM

14.55x

Need 14.55x WSE-3

Etched Sohu

VRAM

4.44x

Need 4.44x Sohu

To match 1x Etched Sohu

NVIDIA A40

Compute (FP32-eq)

133.33x

Need 133.33x A40

FP32 Compute

267.38x

Need 267.38x A40

VRAM

3.00x

Need 3.00x A40

Memory Bandwidth

6.90x

Need 6.90x A40

AMD MI350X

Compute (FP32-eq)

9.52x

Need 9.52x MI350X

VRAM

0.50x

MI350X has 2.00x more

Cerebras WSE-3

VRAM

3.27x

Need 3.27x WSE-3

SambaNova SN40L

VRAM

0.23x

SN40L has 4.44x more

Pricing

Price Type	A40	MI350X	WSE-3	SN40L	Sohu
CAPEX (Street Price)	—	—	—	—	—
OPEX (per hour)	—	—	—	—	—
Price per TFLOPs (FP32-eq)	—	—	—	—	—