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 L4	NVIDIA A10	SambaNova SN40L	NVIDIA RTX 5000 Ada Generation
Architecture	Ada Lovelace	Ampere	Reconfigurable Dataflow Unit (RDU)	Ada Lovelace
Form Factor	PCIe	PCIe	—	PCIe
VRAM	24 GB	24 GB	640 GB	32 GB
Memory Bandwidth	300 GB/s	600 GB/s	—	576 GB/s
TFLOPs (FP32)	30.3	31.2	—	65.3
TFLOPs (FP16)	242	125	—	130.6
TFLOPs	121	125	—	130.6
TFLOPs (FP8)	485	—	—	—
TDP	72 W	150 W	700 W	250 W
Launch Date	Mar 2023	Apr 2021	Jan 2024	Aug 2023

Efficiency Metrics

Metric	L4	A10	SN40L	RTX 5000 Ada Generation
TFLOPs per Watt (FP32-eq)	0.84	0.42	—	0.26
Memory Bandwidth per GB	12.5 GB/s	25.0 GB/s	—	18.0 GB/s

Performance Equivalence

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

To match 1x NVIDIA L4

NVIDIA A10

Compute (FP32-eq)

0.97x

A10 is 1.03x faster

FP32 Compute

0.97x

A10 is 1.03x faster

VRAM

1.00x

A10 has 1.00x more

Memory Bandwidth

0.50x

A10 has 2.00x more

SambaNova SN40L

VRAM

0.04x

SN40L has 26.67x more

NVIDIA RTX 5000 Ada Generation

Compute (FP32-eq)

0.93x

RTX 5000 Ada Generation is 1.08x faster

FP32 Compute

0.46x

RTX 5000 Ada Generation is 2.16x faster

VRAM

0.75x

RTX 5000 Ada Generation has 1.33x more

Memory Bandwidth

0.52x

RTX 5000 Ada Generation has 1.92x more

To match 1x NVIDIA A10

NVIDIA L4

Compute (FP32-eq)

1.03x

Need 1.03x L4

FP32 Compute

1.03x

Need 1.03x L4

VRAM

1.00x

L4 has 1.00x more

Memory Bandwidth

2.00x

Need 2.00x L4

SambaNova SN40L

VRAM

0.04x

SN40L has 26.67x more

NVIDIA RTX 5000 Ada Generation

Compute (FP32-eq)

0.96x

RTX 5000 Ada Generation is 1.04x faster

FP32 Compute

0.48x

RTX 5000 Ada Generation is 2.09x faster

VRAM

0.75x

RTX 5000 Ada Generation has 1.33x more

Memory Bandwidth

1.04x

Need 1.04x RTX 5000 Ada Generation

To match 1x SambaNova SN40L

NVIDIA L4

VRAM

26.67x

Need 26.67x L4

NVIDIA A10

VRAM

26.67x

Need 26.67x A10

NVIDIA RTX 5000 Ada Generation

VRAM

20.00x

Need 20.00x RTX 5000 Ada Generation

To match 1x NVIDIA RTX 5000 Ada Generation

NVIDIA L4

Compute (FP32-eq)

1.08x

Need 1.08x L4

FP32 Compute

2.16x

Need 2.16x L4

VRAM

1.33x

Need 1.33x L4

Memory Bandwidth

1.92x

Need 1.92x L4

NVIDIA A10

Compute (FP32-eq)

1.04x

Need 1.04x A10

FP32 Compute

2.09x

Need 2.09x A10

VRAM

1.33x

Need 1.33x A10

Memory Bandwidth

0.96x

A10 has 1.04x more

SambaNova SN40L

VRAM

0.05x

SN40L has 20.00x more

Pricing

Price Type	L4	A10	SN40L	RTX 5000 Ada Generation
CAPEX (Street Price)	$4,000	—	—	—
OPEX (per hour)	$0.80/hr	—	—	—
Price per TFLOPs (FP32-eq)	$66	—	—	—