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	AMD MI325X	AMD MI350X	NVIDIA RTX 4000 Ada Generation	Huawei Ascend 910B
Architecture	CDNA 3.5	CDNA 4	Ada Lovelace	Da Vinci
Form Factor	OAM	OAM	PCIe	—
VRAM	256 GB	288 GB	20 GB	64 GB
Memory Bandwidth	6,000 GB/s	—	360 GB/s	—
TFLOPs (FP32)	175	—	26.7	—
TFLOPs (FP16)	1,400	—	53.4	640
TFLOPs	1,400	2,100	53.4	—
TFLOPs (FP8)	2,800	—	—	—
TDP	800 W	850 W	130 W	400 W
Launch Date	Oct 2024	Jun 2025	Apr 2023	Aug 2023

Efficiency Metrics

Metric	MI325X	MI350X	RTX 4000 Ada Generation	Ascend 910B
TFLOPs per Watt (FP32-eq)	0.88	1.24	0.21	—
Memory Bandwidth per GB	23.4 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 AMD MI325X

AMD MI350X

Compute (FP32-eq)

0.67x

MI350X is 1.50x faster

VRAM

0.89x

MI350X has 1.13x more

NVIDIA RTX 4000 Ada Generation

Compute (FP32-eq)

26.22x

Need 26.22x RTX 4000 Ada Generation

FP32 Compute

6.55x

Need 6.55x RTX 4000 Ada Generation

VRAM

12.80x

Need 12.80x RTX 4000 Ada Generation

Memory Bandwidth

16.67x

Need 16.67x RTX 4000 Ada Generation

Huawei Ascend 910B

VRAM

4.00x

Need 4.00x Ascend 910B

To match 1x AMD MI350X

AMD MI325X

Compute (FP32-eq)

1.50x

Need 1.50x MI325X

VRAM

1.13x

Need 1.13x MI325X

NVIDIA RTX 4000 Ada Generation

Compute (FP32-eq)

39.33x

Need 39.33x RTX 4000 Ada Generation

VRAM

14.40x

Need 14.40x RTX 4000 Ada Generation

Huawei Ascend 910B

VRAM

4.50x

Need 4.50x Ascend 910B

To match 1x NVIDIA RTX 4000 Ada Generation

AMD MI325X

Compute (FP32-eq)

0.04x

MI325X is 26.22x faster

FP32 Compute

0.15x

MI325X is 6.55x faster

VRAM

0.08x

MI325X has 12.80x more

Memory Bandwidth

0.06x

MI325X has 16.67x more

AMD MI350X

Compute (FP32-eq)

0.03x

MI350X is 39.33x faster

VRAM

0.07x

MI350X has 14.40x more

Huawei Ascend 910B

VRAM

0.31x

Ascend 910B has 3.20x more

To match 1x Huawei Ascend 910B

AMD MI325X

VRAM

0.25x

MI325X has 4.00x more

AMD MI350X

VRAM

0.22x

MI350X has 4.50x more

NVIDIA RTX 4000 Ada Generation

VRAM

3.20x

Need 3.20x RTX 4000 Ada Generation

Pricing

Price Type	MI325X	MI350X	RTX 4000 Ada Generation	Ascend 910B
CAPEX (Street Price)	$45,000	—	—	—
OPEX (per hour)	—	—	—	—
Price per TFLOPs (FP32-eq)	$64	—	—	—