Compare XPUs

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

Select XPUs to Compare

Clear all (3)Jump to results

Filter by Vendor

Showing 128 XPUs • 3 selected

Alibaba

Hanguang 800

AMD

MI100

23.1 TFLOPs

AMD

MI210

181 TFLOPs

AMD

MI250X

383 TFLOPs

AMD

MI300X

1,307 TFLOPs

AMD

MI325X

1,400 TFLOPs

AMD

MI350X

2,100 TFLOPs

AMD

MI355X

5,300 TFLOPs

AMD

Radeon Pro V520

23.04 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

FuriosaAI

RNGD (Renegade)

256 TFLOPs

FuriosaAI

Warboy

Google

TPU v4

275 TFLOPs

Google

TPU v5e

197 TFLOPs

Google

TPU v5p

459 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	Intel Data Center GPU Max 1550	NVIDIA H100 SXM	NVIDIA A100 SXM
Architecture	Ponte Vecchio	Hopper	Ampere
Form Factor	OAM	SXM	SXM
VRAM	128 GB	80 GB	80 GB
Memory Bandwidth	3,277 GB/s	3,350 GB/s	2,039 GB/s
TFLOPs (FP32)	52	67	19.5
TFLOPs (FP16)	419	1,979	312
TFLOPs	419	1,979	312
TFLOPs (FP8)	—	3,958	—
TDP	600 W	700 W	400 W
Launch Date	Jan 2023	Sep 2022	May 2020

Efficiency Metrics

Metric	Data Center GPU Max 1550	H100 SXM	A100 SXM
TFLOPs per Watt (FP32-eq)	0.35	1.41	0.39
Memory Bandwidth per GB	25.6 GB/s	41.9 GB/s	25.5 GB/s

Performance Equivalence

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

To match 1x Intel Data Center GPU Max 1550

NVIDIA H100 SXM

Compute (FP32-eq)

0.21x

H100 SXM is 4.72x faster

FP32 Compute

0.78x

H100 SXM is 1.29x faster

VRAM

1.60x

Need 1.60x H100 SXM

Memory Bandwidth

0.98x

H100 SXM has 1.02x more

NVIDIA A100 SXM

Compute (FP32-eq)

1.34x

Need 1.34x A100 SXM

FP32 Compute

2.67x

Need 2.67x A100 SXM

VRAM

1.60x

Need 1.60x A100 SXM

Memory Bandwidth

1.61x

Need 1.61x A100 SXM

To match 1x NVIDIA H100 SXM

Intel Data Center GPU Max 1550

Compute (FP32-eq)

4.72x

Need 4.72x Data Center GPU Max 1550

FP32 Compute

1.29x

Need 1.29x Data Center GPU Max 1550

VRAM

0.63x

Data Center GPU Max 1550 has 1.60x more

Memory Bandwidth

1.02x

Need 1.02x Data Center GPU Max 1550

NVIDIA A100 SXM

Compute (FP32-eq)

6.34x

Need 6.34x A100 SXM

FP32 Compute

3.44x

Need 3.44x A100 SXM

VRAM

1.00x

A100 SXM has 1.00x more

Memory Bandwidth

1.64x

Need 1.64x A100 SXM

To match 1x NVIDIA A100 SXM

Intel Data Center GPU Max 1550

Compute (FP32-eq)

0.74x

Data Center GPU Max 1550 is 1.34x faster

FP32 Compute

0.38x

Data Center GPU Max 1550 is 2.67x faster

VRAM

0.63x

Data Center GPU Max 1550 has 1.60x more

Memory Bandwidth

0.62x

Data Center GPU Max 1550 has 1.61x more

NVIDIA H100 SXM

Compute (FP32-eq)

0.16x

H100 SXM is 6.34x faster

FP32 Compute

0.29x

H100 SXM is 3.44x faster

VRAM

1.00x

H100 SXM has 1.00x more

Memory Bandwidth

0.61x

H100 SXM has 1.64x more

Pricing

Price Type	Data Center GPU Max 1550	H100 SXM	A100 SXM
CAPEX (Street Price)	$10,000	$30,000	$15,000
OPEX (per hour)	—	$3.50/hr	$4.05/hr
Price per TFLOPs (FP32-eq)	$48	$30	$96