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 A100 SXM	AMD MI325X	NVIDIA GeForce RTX 4070	Huawei Ascend 910B
Architecture	Ampere	CDNA 3.5	Ada Lovelace	Da Vinci
Form Factor	SXM	OAM	PCIe	—
VRAM	80 GB	256 GB	12 GB	64 GB
Memory Bandwidth	2,039 GB/s	6,000 GB/s	504 GB/s	—
TFLOPs (FP32)	19.5	175	29.1	—
TFLOPs (FP16)	312	1,400	58.2	640
TFLOPs	312	1,400	58.2	—
TFLOPs (FP8)	—	2,800	—	—
TDP	400 W	800 W	200 W	400 W
Launch Date	May 2020	Oct 2024	Apr 2023	Aug 2023

Efficiency Metrics

Metric	A100 SXM	MI325X	GeForce RTX 4070	Ascend 910B
TFLOPs per Watt (FP32-eq)	0.39	0.88	0.15	—
Memory Bandwidth per GB	25.5 GB/s	23.4 GB/s	42.0 GB/s	—

Performance Equivalence

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

To match 1x NVIDIA A100 SXM

AMD MI325X

Compute (FP32-eq)

0.22x

MI325X is 4.49x faster

FP32 Compute

0.11x

MI325X is 8.97x faster

VRAM

0.31x

MI325X has 3.20x more

Memory Bandwidth

0.34x

MI325X has 2.94x more

NVIDIA GeForce RTX 4070

Compute (FP32-eq)

5.36x

Need 5.36x GeForce RTX 4070

FP32 Compute

0.67x

GeForce RTX 4070 is 1.49x faster

VRAM

6.67x

Need 6.67x GeForce RTX 4070

Memory Bandwidth

4.05x

Need 4.05x GeForce RTX 4070

Huawei Ascend 910B

VRAM

1.25x

Need 1.25x Ascend 910B

To match 1x AMD MI325X

NVIDIA A100 SXM

Compute (FP32-eq)

4.49x

Need 4.49x A100 SXM

FP32 Compute

8.97x

Need 8.97x A100 SXM

VRAM

3.20x

Need 3.20x A100 SXM

Memory Bandwidth

2.94x

Need 2.94x A100 SXM

NVIDIA GeForce RTX 4070

Compute (FP32-eq)

24.05x

Need 24.05x GeForce RTX 4070

FP32 Compute

6.01x

Need 6.01x GeForce RTX 4070

VRAM

21.33x

Need 21.33x GeForce RTX 4070

Memory Bandwidth

11.90x

Need 11.90x GeForce RTX 4070

Huawei Ascend 910B

VRAM

4.00x

Need 4.00x Ascend 910B

To match 1x NVIDIA GeForce RTX 4070

NVIDIA A100 SXM

Compute (FP32-eq)

0.19x

A100 SXM is 5.36x faster

FP32 Compute

1.49x

Need 1.49x A100 SXM

VRAM

0.15x

A100 SXM has 6.67x more

Memory Bandwidth

0.25x

A100 SXM has 4.05x more

AMD MI325X

Compute (FP32-eq)

0.04x

MI325X is 24.05x faster

FP32 Compute

0.17x

MI325X is 6.01x faster

VRAM

0.05x

MI325X has 21.33x more

Memory Bandwidth

0.08x

MI325X has 11.90x more

Huawei Ascend 910B

VRAM

0.19x

Ascend 910B has 5.33x more

To match 1x Huawei Ascend 910B

NVIDIA A100 SXM

VRAM

0.80x

A100 SXM has 1.25x more

AMD MI325X

VRAM

0.25x

MI325X has 4.00x more

NVIDIA GeForce RTX 4070

VRAM

5.33x

Need 5.33x GeForce RTX 4070

Pricing

Price Type	A100 SXM	MI325X	GeForce RTX 4070	Ascend 910B
CAPEX (Street Price)	$15,000	$45,000	—	—
OPEX (per hour)	$4.05/hr	—	—	—
Price per TFLOPs (FP32-eq)	$96	$64	—	—