Compare XPUs

Select up to 5 XPUs to compare side-by-side

Select XPUs to Compare

Clear all (3)Jump to results

Filter by Vendor

Showing 72 XPUs • 3 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 A40	NVIDIA GB200 NVL72	NVIDIA GeForce RTX 4080
Architecture	Ampere	Blackwell	Ada Lovelace
Form Factor	PCIe	Custom	PCIe
VRAM	48 GB	13824 GB	16 GB
Memory Bandwidth	696 GB/s	130,000 GB/s	717 GB/s
TFLOPs (FP32)	37.4	6,480	48.7
TFLOPs (FP16)	150	360,000	97.5
TFLOPs	150	360,000	97.5
TFLOPs (FP8)	—	—	—
TDP	300 W	120000 W	320 W
Launch Date	Oct 2020	Mar 2024	Nov 2022

Efficiency Metrics

Metric	A40	GB200 NVL72	GeForce RTX 4080
TFLOPs per Watt (FP32-eq)	0.25	1.50	0.15
Memory Bandwidth per GB	14.5 GB/s	9.4 GB/s	44.8 GB/s

Performance Equivalence

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

To match 1x NVIDIA A40

NVIDIA GB200 NVL72

Compute (FP32-eq)

0.00x

GB200 NVL72 is 2400.00x faster

FP32 Compute

0.01x

GB200 NVL72 is 173.26x faster

VRAM

0.00x

GB200 NVL72 has 288.00x more

Memory Bandwidth

0.01x

GB200 NVL72 has 186.78x more

NVIDIA GeForce RTX 4080

Compute (FP32-eq)

1.54x

Need 1.54x GeForce RTX 4080

FP32 Compute

0.77x

GeForce RTX 4080 is 1.30x faster

VRAM

3.00x

Need 3.00x GeForce RTX 4080

Memory Bandwidth

0.97x

GeForce RTX 4080 has 1.03x more

To match 1x NVIDIA GB200 NVL72

NVIDIA A40

Compute (FP32-eq)

2400.00x

Need 2400.00x A40

FP32 Compute

173.26x

Need 173.26x A40

VRAM

288.00x

Need 288.00x A40

Memory Bandwidth

186.78x

Need 186.78x A40

NVIDIA GeForce RTX 4080

Compute (FP32-eq)

3692.31x

Need 3692.31x GeForce RTX 4080

FP32 Compute

133.06x

Need 133.06x GeForce RTX 4080

VRAM

864.00x

Need 864.00x GeForce RTX 4080

Memory Bandwidth

181.31x

Need 181.31x GeForce RTX 4080

To match 1x NVIDIA GeForce RTX 4080

NVIDIA A40

Compute (FP32-eq)

0.65x

A40 is 1.54x faster

FP32 Compute

1.30x

Need 1.30x A40

VRAM

0.33x

A40 has 3.00x more

Memory Bandwidth

1.03x

Need 1.03x A40

NVIDIA GB200 NVL72

Compute (FP32-eq)

0.00x

GB200 NVL72 is 3692.31x faster

FP32 Compute

0.01x

GB200 NVL72 is 133.06x faster

VRAM

0.00x

GB200 NVL72 has 864.00x more

Memory Bandwidth

0.01x

GB200 NVL72 has 181.31x more

Pricing

Price Type	A40	GB200 NVL72	GeForce RTX 4080
CAPEX (Street Price)	—	—	—
OPEX (per hour)	—	—	—
Price per TFLOPs (FP32-eq)	—	—	—