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 L40S	NVIDIA A10	SambaNova SN40L	NVIDIA RTX 6000 Ada Generation
Architecture	Ada Lovelace	Ampere	Reconfigurable Dataflow Unit (RDU)	Ada Lovelace
Form Factor	PCIe	PCIe	—	PCIe
VRAM	48 GB	24 GB	640 GB	48 GB
Memory Bandwidth	864 GB/s	600 GB/s	—	960 GB/s
TFLOPs (FP32)	91.6	31.2	—	91.1
TFLOPs (FP16)	733	125	—	182.2
TFLOPs	733	125	—	182.2
TFLOPs (FP8)	1,466	—	—	—
TDP	350 W	150 W	700 W	300 W
Launch Date	Oct 2023	Apr 2021	Jan 2024	Dec 2022

Efficiency Metrics

Metric	L40S	A10	SN40L	RTX 6000 Ada Generation
TFLOPs per Watt (FP32-eq)	1.05	0.42	—	0.30
Memory Bandwidth per GB	18.0 GB/s	25.0 GB/s	—	20.0 GB/s

Performance Equivalence

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

To match 1x NVIDIA L40S

NVIDIA A10

Compute (FP32-eq)

5.86x

Need 5.86x A10

FP32 Compute

2.94x

Need 2.94x A10

VRAM

2.00x

Need 2.00x A10

Memory Bandwidth

1.44x

Need 1.44x A10

SambaNova SN40L

VRAM

0.07x

SN40L has 13.33x more

NVIDIA RTX 6000 Ada Generation

Compute (FP32-eq)

4.02x

Need 4.02x RTX 6000 Ada Generation

FP32 Compute

1.01x

Need 1.01x RTX 6000 Ada Generation

VRAM

1.00x

RTX 6000 Ada Generation has 1.00x more

Memory Bandwidth

0.90x

RTX 6000 Ada Generation has 1.11x more

To match 1x NVIDIA A10

NVIDIA L40S

Compute (FP32-eq)

0.17x

L40S is 5.86x faster

FP32 Compute

0.34x

L40S is 2.94x faster

VRAM

0.50x

L40S has 2.00x more

Memory Bandwidth

0.69x

L40S has 1.44x more

SambaNova SN40L

VRAM

0.04x

SN40L has 26.67x more

NVIDIA RTX 6000 Ada Generation

Compute (FP32-eq)

0.69x

RTX 6000 Ada Generation is 1.46x faster

FP32 Compute

0.34x

RTX 6000 Ada Generation is 2.92x faster

VRAM

0.50x

RTX 6000 Ada Generation has 2.00x more

Memory Bandwidth

0.63x

RTX 6000 Ada Generation has 1.60x more

To match 1x SambaNova SN40L

NVIDIA L40S

VRAM

13.33x

Need 13.33x L40S

NVIDIA A10

VRAM

26.67x

Need 26.67x A10

NVIDIA RTX 6000 Ada Generation

VRAM

13.33x

Need 13.33x RTX 6000 Ada Generation

To match 1x NVIDIA RTX 6000 Ada Generation

NVIDIA L40S

Compute (FP32-eq)

0.25x

L40S is 4.02x faster

FP32 Compute

0.99x

L40S is 1.01x faster

VRAM

1.00x

L40S has 1.00x more

Memory Bandwidth

1.11x

Need 1.11x L40S

NVIDIA A10

Compute (FP32-eq)

1.46x

Need 1.46x A10

FP32 Compute

2.92x

Need 2.92x A10

VRAM

2.00x

Need 2.00x A10

Memory Bandwidth

1.60x

Need 1.60x A10

SambaNova SN40L

VRAM

0.07x

SN40L has 13.33x more

Pricing

Price Type	L40S	A10	SN40L	RTX 6000 Ada Generation
CAPEX (Street Price)	$10,000	—	—	—
OPEX (per hour)	$1.50/hr	—	—	—
Price per TFLOPs (FP32-eq)	$27	—	—	—