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 B200	Cerebras WSE-3	AWS Inferentia2	Meta MTIA v1	AMD MI300A
Architecture	Blackwell	Wafer Scale Engine	Inferentia Gen2	Meta Training & Inference Accelerator	CDNA 3
Form Factor	SXM	—	—	—	OAM
VRAM	192 GB	44 GB	32 GB	128 GB	128 GB
Memory Bandwidth	8,000 GB/s	—	—	—	5,300 GB/s
TFLOPs (FP32)	90	—	—	—	122.6
TFLOPs (FP16)	2,250	—	—	—	980.6
TFLOPs	2,250	—	190	—	980.6
TFLOPs (FP8)	4,500	—	—	—	—
TDP	1000 W	23000 W	150 W	400 W	760 W
Launch Date	Mar 2024	Mar 2024	Nov 2022	May 2023	Dec 2023

Efficiency Metrics

Metric	B200	WSE-3	Inferentia2	MTIA v1	MI300A
TFLOPs per Watt (FP32-eq)	1.13	—	0.63	—	0.65
Memory Bandwidth per GB	41.7 GB/s	—	—	—	41.4 GB/s

Performance Equivalence

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

To match 1x NVIDIA B200

Cerebras WSE-3

VRAM

4.36x

Need 4.36x WSE-3

AWS Inferentia2

Compute (FP32-eq)

11.84x

Need 11.84x Inferentia2

VRAM

6.00x

Need 6.00x Inferentia2

Meta MTIA v1

VRAM

1.50x

Need 1.50x MTIA v1

AMD MI300A

Compute (FP32-eq)

2.29x

Need 2.29x MI300A

FP32 Compute

0.73x

MI300A is 1.36x faster

VRAM

1.50x

Need 1.50x MI300A

Memory Bandwidth

1.51x

Need 1.51x MI300A

To match 1x Cerebras WSE-3

NVIDIA B200

VRAM

0.23x

B200 has 4.36x more

AWS Inferentia2

VRAM

1.38x

Need 1.38x Inferentia2

Meta MTIA v1

VRAM

0.34x

MTIA v1 has 2.91x more

AMD MI300A

VRAM

0.34x

MI300A has 2.91x more

To match 1x AWS Inferentia2

NVIDIA B200

Compute (FP32-eq)

0.08x

B200 is 11.84x faster

VRAM

0.17x

B200 has 6.00x more

Cerebras WSE-3

VRAM

0.73x

WSE-3 has 1.38x more

Meta MTIA v1

VRAM

0.25x

MTIA v1 has 4.00x more

AMD MI300A

Compute (FP32-eq)

0.19x

MI300A is 5.16x faster

VRAM

0.25x

MI300A has 4.00x more

To match 1x Meta MTIA v1

NVIDIA B200

VRAM

0.67x

B200 has 1.50x more

Cerebras WSE-3

VRAM

2.91x

Need 2.91x WSE-3

AWS Inferentia2

VRAM

4.00x

Need 4.00x Inferentia2

AMD MI300A

VRAM

1.00x

MI300A has 1.00x more

To match 1x AMD MI300A

NVIDIA B200

Compute (FP32-eq)

0.44x

B200 is 2.29x faster

FP32 Compute

1.36x

Need 1.36x B200

VRAM

0.67x

B200 has 1.50x more

Memory Bandwidth

0.66x

B200 has 1.51x more

Cerebras WSE-3

VRAM

2.91x

Need 2.91x WSE-3

AWS Inferentia2

Compute (FP32-eq)

5.16x

Need 5.16x Inferentia2

VRAM

4.00x

Need 4.00x Inferentia2

Meta MTIA v1

VRAM

1.00x

MTIA v1 has 1.00x more

Pricing

Price Type	B200	WSE-3	Inferentia2	MTIA v1	MI300A
CAPEX (Street Price)	$70,000	—	—	—	—
OPEX (per hour)	—	—	—	—	—
Price per TFLOPs (FP32-eq)	$62	—	—	—	—