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	Graphcore Bow IPU	Google TPU v5p	NVIDIA GeForce GT 710	NVIDIA RTX A4000	NVIDIA A100 SXM
Architecture	Bow-2000	TPU v5	Kepler	Ampere	Ampere
Form Factor	—	Mezzanine	PCIe	PCIe	SXM
VRAM	3.6 GB	95 GB	2 GB	16 GB	80 GB
Memory Bandwidth	—	—	14.4 GB/s	448 GB/s	2,039 GB/s
TFLOPs (FP32)	—	—	0.366	19.2	19.5
TFLOPs (FP16)	1,000	—	—	—	312
TFLOPs	—	459	0.366	38.7	312
TFLOPs (FP8)	—	—	—	—	—
TDP	200 W	400 W	19 W	140 W	400 W
Launch Date	Mar 2022	Aug 2023	Mar 2014	Apr 2021	May 2020

Efficiency Metrics

Metric	Bow IPU	TPU v5p	GeForce GT 710	RTX A4000	A100 SXM
TFLOPs per Watt (FP32-eq)	—	0.57	0.02	0.14	0.39
Memory Bandwidth per GB	—	—	7.2 GB/s	28.0 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 Graphcore Bow IPU

Google TPU v5p

VRAM

0.04x

TPU v5p has 26.39x more

NVIDIA GeForce GT 710

VRAM

1.80x

Need 1.80x GeForce GT 710

NVIDIA RTX A4000

VRAM

0.23x

RTX A4000 has 4.44x more

NVIDIA A100 SXM

VRAM

0.04x

A100 SXM has 22.22x more

To match 1x Google TPU v5p

Graphcore Bow IPU

VRAM

26.39x

Need 26.39x Bow IPU

NVIDIA GeForce GT 710

Compute (FP32-eq)

627.05x

Need 627.05x GeForce GT 710

VRAM

47.50x

Need 47.50x GeForce GT 710

NVIDIA RTX A4000

Compute (FP32-eq)

11.86x

Need 11.86x RTX A4000

VRAM

5.94x

Need 5.94x RTX A4000

NVIDIA A100 SXM

Compute (FP32-eq)

1.47x

Need 1.47x A100 SXM

VRAM

1.19x

Need 1.19x A100 SXM

To match 1x NVIDIA GeForce GT 710

Graphcore Bow IPU

VRAM

0.56x

Bow IPU has 1.80x more

Google TPU v5p

Compute (FP32-eq)

0.00x

TPU v5p is 627.05x faster

VRAM

0.02x

TPU v5p has 47.50x more

NVIDIA RTX A4000

Compute (FP32-eq)

0.02x

RTX A4000 is 52.87x faster

FP32 Compute

0.02x

RTX A4000 is 52.46x faster

VRAM

0.13x

RTX A4000 has 8.00x more

Memory Bandwidth

0.03x

RTX A4000 has 31.11x more

NVIDIA A100 SXM

Compute (FP32-eq)

0.00x

A100 SXM is 426.23x faster

FP32 Compute

0.02x

A100 SXM is 53.28x faster

VRAM

0.03x

A100 SXM has 40.00x more

Memory Bandwidth

0.01x

A100 SXM has 141.60x more

To match 1x NVIDIA RTX A4000

Graphcore Bow IPU

VRAM

4.44x

Need 4.44x Bow IPU

Google TPU v5p

Compute (FP32-eq)

0.08x

TPU v5p is 11.86x faster

VRAM

0.17x

TPU v5p has 5.94x more

NVIDIA GeForce GT 710

Compute (FP32-eq)

52.87x

Need 52.87x GeForce GT 710

FP32 Compute

52.46x

Need 52.46x GeForce GT 710

VRAM

8.00x

Need 8.00x GeForce GT 710

Memory Bandwidth

31.11x

Need 31.11x GeForce GT 710

NVIDIA A100 SXM

Compute (FP32-eq)

0.12x

A100 SXM is 8.06x faster

FP32 Compute

0.98x

A100 SXM is 1.02x faster

VRAM

0.20x

A100 SXM has 5.00x more

Memory Bandwidth

0.22x

A100 SXM has 4.55x more

To match 1x NVIDIA A100 SXM

Graphcore Bow IPU

VRAM

22.22x

Need 22.22x Bow IPU

Google TPU v5p

Compute (FP32-eq)

0.68x

TPU v5p is 1.47x faster

VRAM

0.84x

TPU v5p has 1.19x more

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

141.60x

Need 141.60x GeForce GT 710

NVIDIA RTX A4000

Compute (FP32-eq)

8.06x

Need 8.06x RTX A4000

FP32 Compute

1.02x

Need 1.02x RTX A4000

VRAM

5.00x

Need 5.00x RTX A4000

Memory Bandwidth

4.55x

Need 4.55x RTX A4000

Pricing

Price Type	Bow IPU	TPU v5p	GeForce GT 710	RTX A4000	A100 SXM
CAPEX (Street Price)	—	—	—	—	$15,000
OPEX (per hour)	—	$5.00/hr	$0.07/hr	$0.11/hr	$4.05/hr
Price per TFLOPs (FP32-eq)	—	—	—	—	$96