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 H200 SXM	AMD MI350X	AWS Trainium2	NVIDIA GeForce RTX 4080
Architecture	Hopper	CDNA 4	Trainium Gen2	Ada Lovelace
Form Factor	SXM	OAM	—	PCIe
VRAM	141 GB	288 GB	—	16 GB
Memory Bandwidth	4,800 GB/s	—	—	717 GB/s
TFLOPs (FP32)	67	—	—	48.7
TFLOPs (FP16)	1,979	—	—	97.5
TFLOPs	1,979	2,100	680	97.5
TFLOPs (FP8)	3,958	—	—	—
TDP	700 W	850 W	300 W	320 W
Launch Date	Nov 2023	Jun 2025	Nov 2023	Nov 2022

Efficiency Metrics

Metric	H200 SXM	MI350X	Trainium2	GeForce RTX 4080
TFLOPs per Watt (FP32-eq)	1.41	1.24	1.13	0.15
Memory Bandwidth per GB	34.0 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 H200 SXM

AMD MI350X

Compute (FP32-eq)

0.94x

MI350X is 1.06x faster

VRAM

0.49x

MI350X has 2.04x more

AWS Trainium2

Compute (FP32-eq)

2.91x

Need 2.91x Trainium2

NVIDIA GeForce RTX 4080

Compute (FP32-eq)

20.30x

Need 20.30x GeForce RTX 4080

FP32 Compute

1.38x

Need 1.38x GeForce RTX 4080

VRAM

8.81x

Need 8.81x GeForce RTX 4080

Memory Bandwidth

6.69x

Need 6.69x GeForce RTX 4080

To match 1x AMD MI350X

NVIDIA H200 SXM

Compute (FP32-eq)

1.06x

Need 1.06x H200 SXM

VRAM

2.04x

Need 2.04x H200 SXM

AWS Trainium2

Compute (FP32-eq)

3.09x

Need 3.09x Trainium2

NVIDIA GeForce RTX 4080

Compute (FP32-eq)

21.54x

Need 21.54x GeForce RTX 4080

VRAM

18.00x

Need 18.00x GeForce RTX 4080

To match 1x AWS Trainium2

NVIDIA H200 SXM

Compute (FP32-eq)

0.34x

H200 SXM is 2.91x faster

AMD MI350X

Compute (FP32-eq)

0.32x

MI350X is 3.09x faster

NVIDIA GeForce RTX 4080

Compute (FP32-eq)

6.97x

Need 6.97x GeForce RTX 4080

To match 1x NVIDIA GeForce RTX 4080

NVIDIA H200 SXM

Compute (FP32-eq)

0.05x

H200 SXM is 20.30x faster

FP32 Compute

0.73x

H200 SXM is 1.38x faster

VRAM

0.11x

H200 SXM has 8.81x more

Memory Bandwidth

0.15x

H200 SXM has 6.69x more

AMD MI350X

Compute (FP32-eq)

0.05x

MI350X is 21.54x faster

VRAM

0.06x

MI350X has 18.00x more

AWS Trainium2

Compute (FP32-eq)

0.14x

Trainium2 is 6.97x faster

Pricing

Price Type	H200 SXM	MI350X	Trainium2	GeForce RTX 4080
CAPEX (Street Price)	$40,000	—	—	—
OPEX (per hour)	—	—	$1.34/hr	—
Price per TFLOPs (FP32-eq)	$40	—	—	—