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	AMD MI210	AMD MI325X	AWS Inferentia2
Architecture	Ada Lovelace	CDNA 2	CDNA 3.5	Inferentia Gen2
Form Factor	PCIe	PCIe	OAM	—
VRAM	48 GB	64 GB	256 GB	32 GB
Memory Bandwidth	864 GB/s	1,638 GB/s	6,000 GB/s	—
TFLOPs (FP32)	91.6	45.3	175	—
TFLOPs (FP16)	733	181	1,400	—
TFLOPs	733	181	1,400	190
TFLOPs (FP8)	1,466	—	2,800	—
TDP	350 W	300 W	800 W	150 W
Launch Date	Oct 2023	Jan 2022	Oct 2024	Nov 2022

Efficiency Metrics

Metric	L40S	MI210	MI325X	Inferentia2
TFLOPs per Watt (FP32-eq)	1.05	0.30	0.88	0.63
Memory Bandwidth per GB	18.0 GB/s	25.6 GB/s	23.4 GB/s	—

Performance Equivalence

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

To match 1x NVIDIA L40S

AMD MI210

Compute (FP32-eq)

4.05x

Need 4.05x MI210

FP32 Compute

2.02x

Need 2.02x MI210

VRAM

0.75x

MI210 has 1.33x more

Memory Bandwidth

0.53x

MI210 has 1.90x more

AMD MI325X

Compute (FP32-eq)

0.52x

MI325X is 1.91x faster

FP32 Compute

0.52x

MI325X is 1.91x faster

VRAM

0.19x

MI325X has 5.33x more

Memory Bandwidth

0.14x

MI325X has 6.94x more

AWS Inferentia2

Compute (FP32-eq)

3.86x

Need 3.86x Inferentia2

VRAM

1.50x

Need 1.50x Inferentia2

To match 1x AMD MI210

NVIDIA L40S

Compute (FP32-eq)

0.25x

L40S is 4.05x faster

FP32 Compute

0.49x

L40S is 2.02x faster

VRAM

1.33x

Need 1.33x L40S

Memory Bandwidth

1.90x

Need 1.90x L40S

AMD MI325X

Compute (FP32-eq)

0.13x

MI325X is 7.73x faster

FP32 Compute

0.26x

MI325X is 3.86x faster

VRAM

0.25x

MI325X has 4.00x more

Memory Bandwidth

0.27x

MI325X has 3.66x more

AWS Inferentia2

Compute (FP32-eq)

0.95x

Inferentia2 is 1.05x faster

VRAM

2.00x

Need 2.00x Inferentia2

To match 1x AMD MI325X

NVIDIA L40S

Compute (FP32-eq)

1.91x

Need 1.91x L40S

FP32 Compute

1.91x

Need 1.91x L40S

VRAM

5.33x

Need 5.33x L40S

Memory Bandwidth

6.94x

Need 6.94x L40S

AMD MI210

Compute (FP32-eq)

7.73x

Need 7.73x MI210

FP32 Compute

3.86x

Need 3.86x MI210

VRAM

4.00x

Need 4.00x MI210

Memory Bandwidth

3.66x

Need 3.66x MI210

AWS Inferentia2

Compute (FP32-eq)

7.37x

Need 7.37x Inferentia2

VRAM

8.00x

Need 8.00x Inferentia2

To match 1x AWS Inferentia2

NVIDIA L40S

Compute (FP32-eq)

0.26x

L40S is 3.86x faster

VRAM

0.67x

L40S has 1.50x more

AMD MI210

Compute (FP32-eq)

1.05x

Need 1.05x MI210

VRAM

0.50x

MI210 has 2.00x more

AMD MI325X

Compute (FP32-eq)

0.14x

MI325X is 7.37x faster

VRAM

0.13x

MI325X has 8.00x more

Pricing

Price Type	L40S	MI210	MI325X	Inferentia2
CAPEX (Street Price)	$10,000	$6,000	$45,000	—
OPEX (per hour)	$1.50/hr	—	—	—
Price per TFLOPs (FP32-eq)	$27	$66	$64	—