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 B200	Tenstorrent Grayskull	AWS Inferentia2	Iluvatar CoreX BI-V150
Architecture	Blackwell	Tensix Core	Inferentia Gen2	Iluvatar GPU
Form Factor	SXM	—	—	—
VRAM	192 GB	8 GB	32 GB	48 GB
Memory Bandwidth	8,000 GB/s	—	—	—
TFLOPs (FP32)	90	—	—	—
TFLOPs (FP16)	2,250	—	—	—
TFLOPs	2,250	200	190	300
TFLOPs (FP8)	4,500	—	—	—
TDP	1000 W	65 W	150 W	350 W
Launch Date	Mar 2024	Mar 2021	Nov 2022	Apr 2023

Efficiency Metrics

Metric	B200	Grayskull	Inferentia2	BI-V150
TFLOPs per Watt (FP32-eq)	1.13	1.54	0.63	0.43
Memory Bandwidth per GB	41.7 GB/s	—	—	—

Performance Equivalence

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

To match 1x NVIDIA B200

Tenstorrent Grayskull

Compute (FP32-eq)

11.25x

Need 11.25x Grayskull

VRAM

24.00x

Need 24.00x Grayskull

AWS Inferentia2

Compute (FP32-eq)

11.84x

Need 11.84x Inferentia2

VRAM

6.00x

Need 6.00x Inferentia2

Iluvatar CoreX BI-V150

Compute (FP32-eq)

7.50x

Need 7.50x BI-V150

VRAM

4.00x

Need 4.00x BI-V150

To match 1x Tenstorrent Grayskull

NVIDIA B200

Compute (FP32-eq)

0.09x

B200 is 11.25x faster

VRAM

0.04x

B200 has 24.00x more

AWS Inferentia2

Compute (FP32-eq)

1.05x

Need 1.05x Inferentia2

VRAM

0.25x

Inferentia2 has 4.00x more

Iluvatar CoreX BI-V150

Compute (FP32-eq)

0.67x

BI-V150 is 1.50x faster

VRAM

0.17x

BI-V150 has 6.00x 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

Tenstorrent Grayskull

Compute (FP32-eq)

0.95x

Grayskull is 1.05x faster

VRAM

4.00x

Need 4.00x Grayskull

Iluvatar CoreX BI-V150

Compute (FP32-eq)

0.63x

BI-V150 is 1.58x faster

VRAM

0.67x

BI-V150 has 1.50x more

To match 1x Iluvatar CoreX BI-V150

NVIDIA B200

Compute (FP32-eq)

0.13x

B200 is 7.50x faster

VRAM

0.25x

B200 has 4.00x more

Tenstorrent Grayskull

Compute (FP32-eq)

1.50x

Need 1.50x Grayskull

VRAM

6.00x

Need 6.00x Grayskull

AWS Inferentia2

Compute (FP32-eq)

1.58x

Need 1.58x Inferentia2

VRAM

1.50x

Need 1.50x Inferentia2

Pricing

Price Type	B200	Grayskull	Inferentia2	BI-V150
CAPEX (Street Price)	$70,000	—	—	—
OPEX (per hour)	—	—	—	—
Price per TFLOPs (FP32-eq)	$62	—	—	—