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 128 XPUs • 4 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	SambaNova SN40L	NVIDIA Quadro P1000	NVIDIA GeForce GT 710	NVIDIA GH200
Architecture	Reconfigurable Dataflow Unit (RDU)	Pascal	Kepler	Hopper
Form Factor	—	PCIe	PCIe	SXM
VRAM	640 GB	4 GB	2 GB	96 GB
Memory Bandwidth	—	82 GB/s	14.4 GB/s	4,000 GB/s
TFLOPs (FP32)	—	1.894	0.366	67
TFLOPs (FP16)	—	—	—	—
TFLOPs	—	1.894	0.366	989
TFLOPs (FP8)	—	—	—	—
TDP	700 W	47 W	19 W	1000 W
Launch Date	Jan 2024	Feb 2017	Mar 2014	May 2023

Efficiency Metrics

Metric	SN40L	Quadro P1000	GeForce GT 710	GH200
TFLOPs per Watt (FP32-eq)	—	0.04	0.02	0.49
Memory Bandwidth per GB	—	20.5 GB/s	7.2 GB/s	41.7 GB/s

Performance Equivalence

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

To match 1x SambaNova SN40L

NVIDIA Quadro P1000

VRAM

160.00x

Need 160.00x Quadro P1000

NVIDIA GeForce GT 710

VRAM

320.00x

Need 320.00x GeForce GT 710

NVIDIA GH200

VRAM

6.67x

Need 6.67x GH200

To match 1x NVIDIA Quadro P1000

SambaNova SN40L

VRAM

0.01x

SN40L has 160.00x more

NVIDIA GeForce GT 710

Compute (FP32-eq)

5.17x

Need 5.17x GeForce GT 710

FP32 Compute

5.17x

Need 5.17x GeForce GT 710

VRAM

2.00x

Need 2.00x GeForce GT 710

Memory Bandwidth

5.69x

Need 5.69x GeForce GT 710

NVIDIA GH200

Compute (FP32-eq)

0.00x

GH200 is 261.09x faster

FP32 Compute

0.03x

GH200 is 35.37x faster

VRAM

0.04x

GH200 has 24.00x more

Memory Bandwidth

0.02x

GH200 has 48.78x more

To match 1x NVIDIA GeForce GT 710

SambaNova SN40L

VRAM

0.00x

SN40L has 320.00x more

NVIDIA Quadro P1000

Compute (FP32-eq)

0.19x

Quadro P1000 is 5.17x faster

FP32 Compute

0.19x

Quadro P1000 is 5.17x faster

VRAM

0.50x

Quadro P1000 has 2.00x more

Memory Bandwidth

0.18x

Quadro P1000 has 5.69x more

NVIDIA GH200

Compute (FP32-eq)

0.00x

GH200 is 1351.09x faster

FP32 Compute

0.01x

GH200 is 183.06x faster

VRAM

0.02x

GH200 has 48.00x more

Memory Bandwidth

0.00x

GH200 has 277.78x more

To match 1x NVIDIA GH200

SambaNova SN40L

VRAM

0.15x

SN40L has 6.67x more

NVIDIA Quadro P1000

Compute (FP32-eq)

261.09x

Need 261.09x Quadro P1000

FP32 Compute

35.37x

Need 35.37x Quadro P1000

VRAM

24.00x

Need 24.00x Quadro P1000

Memory Bandwidth

48.78x

Need 48.78x Quadro P1000

NVIDIA GeForce GT 710

Compute (FP32-eq)

1351.09x

Need 1351.09x GeForce GT 710

FP32 Compute

183.06x

Need 183.06x GeForce GT 710

VRAM

48.00x

Need 48.00x GeForce GT 710

Memory Bandwidth

277.78x

Need 277.78x GeForce GT 710

Pricing

Price Type	SN40L	Quadro P1000	GeForce GT 710	GH200
CAPEX (Street Price)	—	—	—	—
OPEX (per hour)	—	$0.10/hr	$0.07/hr	$1.49/hr
Price per TFLOPs (FP32-eq)	—	—	—	—