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 128 XPUs • 5 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 Habana Gaudi 2	Intel Habana Gaudi 3	Intel Data Center GPU Max 1100	NVIDIA A800	NVIDIA GeForce GT 710
Architecture	Gaudi Gen2	Gaudi Gen3	Ponte Vecchio	Ampere	Kepler
Form Factor	OAM	OAM	OAM	SXM	PCIe
VRAM	96 GB	128 GB	48 GB	80 GB	2 GB
Memory Bandwidth	2,450 GB/s	3,700 GB/s	1,229 GB/s	1,935 GB/s	14.4 GB/s
TFLOPs (FP32)	—	—	22	19.5	0.366
TFLOPs (FP16)	—	—	177	—	—
TFLOPs	432	1,835	177	312	0.366
TFLOPs (FP8)	—	3,670	—	—	—
TDP	600 W	900 W	300 W	400 W	19 W
Launch Date	May 2022	Apr 2024	Jan 2023	Nov 2022	Mar 2014

Efficiency Metrics

Metric	Gaudi 2	Gaudi 3	Data Center GPU Max 1100	A800	GeForce GT 710
TFLOPs per Watt (FP32-eq)	0.36	1.02	0.29	0.39	0.02
Memory Bandwidth per GB	25.5 GB/s	28.9 GB/s	25.6 GB/s	24.2 GB/s	7.2 GB/s

Performance Equivalence

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

To match 1x Intel Habana Gaudi 2

Intel Habana Gaudi 3

Compute (FP32-eq)

0.24x

Gaudi 3 is 4.25x faster

VRAM

0.75x

Gaudi 3 has 1.33x more

Memory Bandwidth

0.66x

Gaudi 3 has 1.51x more

Intel Data Center GPU Max 1100

Compute (FP32-eq)

2.44x

Need 2.44x Data Center GPU Max 1100

VRAM

2.00x

Need 2.00x Data Center GPU Max 1100

Memory Bandwidth

1.99x

Need 1.99x Data Center GPU Max 1100

NVIDIA A800

Compute (FP32-eq)

1.38x

Need 1.38x A800

VRAM

1.20x

Need 1.20x A800

Memory Bandwidth

1.27x

Need 1.27x A800

NVIDIA GeForce GT 710

Compute (FP32-eq)

590.16x

Need 590.16x GeForce GT 710

VRAM

48.00x

Need 48.00x GeForce GT 710

Memory Bandwidth

170.14x

Need 170.14x GeForce GT 710

To match 1x Intel Habana Gaudi 3

Intel Habana Gaudi 2

Compute (FP32-eq)

4.25x

Need 4.25x Gaudi 2

VRAM

1.33x

Need 1.33x Gaudi 2

Memory Bandwidth

1.51x

Need 1.51x Gaudi 2

Intel Data Center GPU Max 1100

Compute (FP32-eq)

10.37x

Need 10.37x Data Center GPU Max 1100

VRAM

2.67x

Need 2.67x Data Center GPU Max 1100

Memory Bandwidth

3.01x

Need 3.01x Data Center GPU Max 1100

NVIDIA A800

Compute (FP32-eq)

5.88x

Need 5.88x A800

VRAM

1.60x

Need 1.60x A800

Memory Bandwidth

1.91x

Need 1.91x A800

NVIDIA GeForce GT 710

Compute (FP32-eq)

2506.83x

Need 2506.83x GeForce GT 710

VRAM

64.00x

Need 64.00x GeForce GT 710

Memory Bandwidth

256.94x

Need 256.94x GeForce GT 710

To match 1x Intel Data Center GPU Max 1100

Intel Habana Gaudi 2

Compute (FP32-eq)

0.41x

Gaudi 2 is 2.44x faster

VRAM

0.50x

Gaudi 2 has 2.00x more

Memory Bandwidth

0.50x

Gaudi 2 has 1.99x more

Intel Habana Gaudi 3

Compute (FP32-eq)

0.10x

Gaudi 3 is 10.37x faster

VRAM

0.38x

Gaudi 3 has 2.67x more

Memory Bandwidth

0.33x

Gaudi 3 has 3.01x more

NVIDIA A800

Compute (FP32-eq)

0.57x

A800 is 1.76x faster

FP32 Compute

1.13x

Need 1.13x A800

VRAM

0.60x

A800 has 1.67x more

Memory Bandwidth

0.64x

A800 has 1.57x more

NVIDIA GeForce GT 710

Compute (FP32-eq)

241.80x

Need 241.80x GeForce GT 710

FP32 Compute

60.11x

Need 60.11x GeForce GT 710

VRAM

24.00x

Need 24.00x GeForce GT 710

Memory Bandwidth

85.35x

Need 85.35x GeForce GT 710

To match 1x NVIDIA A800

Intel Habana Gaudi 2

Compute (FP32-eq)

0.72x

Gaudi 2 is 1.38x faster

VRAM

0.83x

Gaudi 2 has 1.20x more

Memory Bandwidth

0.79x

Gaudi 2 has 1.27x more

Intel Habana Gaudi 3

Compute (FP32-eq)

0.17x

Gaudi 3 is 5.88x faster

VRAM

0.63x

Gaudi 3 has 1.60x more

Memory Bandwidth

0.52x

Gaudi 3 has 1.91x more

Intel Data Center GPU Max 1100

Compute (FP32-eq)

1.76x

Need 1.76x Data Center GPU Max 1100

FP32 Compute

0.89x

Data Center GPU Max 1100 is 1.13x faster

VRAM

1.67x

Need 1.67x Data Center GPU Max 1100

Memory Bandwidth

1.57x

Need 1.57x Data Center GPU Max 1100

NVIDIA GeForce GT 710

Compute (FP32-eq)

426.23x

Need 426.23x GeForce GT 710

FP32 Compute

53.28x

Need 53.28x GeForce GT 710

VRAM

40.00x

Need 40.00x GeForce GT 710

Memory Bandwidth

134.38x

Need 134.38x GeForce GT 710

To match 1x NVIDIA GeForce GT 710

Intel Habana Gaudi 2

Compute (FP32-eq)

0.00x

Gaudi 2 is 590.16x faster

VRAM

0.02x

Gaudi 2 has 48.00x more

Memory Bandwidth

0.01x

Gaudi 2 has 170.14x more

Intel Habana Gaudi 3

Compute (FP32-eq)

0.00x

Gaudi 3 is 2506.83x faster

VRAM

0.02x

Gaudi 3 has 64.00x more

Memory Bandwidth

0.00x

Gaudi 3 has 256.94x more

Intel Data Center GPU Max 1100

Compute (FP32-eq)

0.00x

Data Center GPU Max 1100 is 241.80x faster

FP32 Compute

0.02x

Data Center GPU Max 1100 is 60.11x faster

VRAM

0.04x

Data Center GPU Max 1100 has 24.00x more

Memory Bandwidth

0.01x

Data Center GPU Max 1100 has 85.35x more

NVIDIA A800

Compute (FP32-eq)

0.00x

A800 is 426.23x faster

FP32 Compute

0.02x

A800 is 53.28x faster

VRAM

0.03x

A800 has 40.00x more

Memory Bandwidth

0.01x

A800 has 134.38x more

Pricing

Price Type	Gaudi 2	Gaudi 3	Data Center GPU Max 1100	A800	GeForce GT 710
CAPEX (Street Price)	—	$15,000	$5,000	—	—
OPEX (per hour)	—	$1.20/hr	—	$0.80/hr	$0.07/hr
Price per TFLOPs (FP32-eq)	—	$16	$56	—	—