Compare XPUs

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

Select XPUs to Compare

Clear all (6)Jump to results

Filter by Vendor

Showing 72 XPUs • 6 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	AMD MI210	AMD MI350X	Tenstorrent Grayskull	FuriosaAI Warboy	AMD MI300A
Architecture	Ampere	CDNA 2	CDNA 4	Tensix Core	FuriosaAI NPU	CDNA 3
Form Factor	PCIe	PCIe	OAM	—	—	OAM
VRAM	48 GB	64 GB	288 GB	8 GB	32 GB	128 GB
Memory Bandwidth	696 GB/s	1,638 GB/s	—	—	—	5,300 GB/s
TFLOPs (FP32)	37.4	45.3	—	—	—	122.6
TFLOPs (FP16)	150	181	—	—	—	980.6
TFLOPs	150	181	2,100	200	—	980.6
TFLOPs (FP8)	—	—	—	—	—	—
TDP	300 W	300 W	850 W	65 W	80 W	760 W
Launch Date	Oct 2020	Jan 2022	Jun 2025	Mar 2021	Oct 2021	Dec 2023

Efficiency Metrics

Metric	A40	MI210	MI350X	Grayskull	Warboy	MI300A
TFLOPs per Watt (FP32-eq)	0.25	0.30	1.24	1.54	—	0.65
Memory Bandwidth per GB	14.5 GB/s	25.6 GB/s	—	—	—	41.4 GB/s

Performance Equivalence

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

To match 1x NVIDIA A40

AMD MI210

Compute (FP32-eq)

0.83x

MI210 is 1.21x faster

FP32 Compute

0.83x

MI210 is 1.21x faster

VRAM

0.75x

MI210 has 1.33x more

Memory Bandwidth

0.42x

MI210 has 2.35x more

AMD MI350X

Compute (FP32-eq)

0.07x

MI350X is 14.00x faster

VRAM

0.17x

MI350X has 6.00x more

Tenstorrent Grayskull

Compute (FP32-eq)

0.75x

Grayskull is 1.33x faster

VRAM

6.00x

Need 6.00x Grayskull

FuriosaAI Warboy

VRAM

1.50x

Need 1.50x Warboy

AMD MI300A

Compute (FP32-eq)

0.15x

MI300A is 6.54x faster

FP32 Compute

0.31x

MI300A is 3.28x faster

VRAM

0.38x

MI300A has 2.67x more

Memory Bandwidth

0.13x

MI300A has 7.61x more

To match 1x AMD MI210

NVIDIA A40

Compute (FP32-eq)

1.21x

Need 1.21x A40

FP32 Compute

1.21x

Need 1.21x A40

VRAM

1.33x

Need 1.33x A40

Memory Bandwidth

2.35x

Need 2.35x A40

AMD MI350X

Compute (FP32-eq)

0.09x

MI350X is 11.60x faster

VRAM

0.22x

MI350X has 4.50x more

Tenstorrent Grayskull

Compute (FP32-eq)

0.91x

Grayskull is 1.10x faster

VRAM

8.00x

Need 8.00x Grayskull

FuriosaAI Warboy

VRAM

2.00x

Need 2.00x Warboy

AMD MI300A

Compute (FP32-eq)

0.18x

MI300A is 5.42x faster

FP32 Compute

0.37x

MI300A is 2.71x faster

VRAM

0.50x

MI300A has 2.00x more

Memory Bandwidth

0.31x

MI300A has 3.24x more

To match 1x AMD MI350X

NVIDIA A40

Compute (FP32-eq)

14.00x

Need 14.00x A40

VRAM

6.00x

Need 6.00x A40

AMD MI210

Compute (FP32-eq)

11.60x

Need 11.60x MI210

VRAM

4.50x

Need 4.50x MI210

Tenstorrent Grayskull

Compute (FP32-eq)

10.50x

Need 10.50x Grayskull

VRAM

36.00x

Need 36.00x Grayskull

FuriosaAI Warboy

VRAM

9.00x

Need 9.00x Warboy

AMD MI300A

Compute (FP32-eq)

2.14x

Need 2.14x MI300A

VRAM

2.25x

Need 2.25x MI300A

To match 1x Tenstorrent Grayskull

NVIDIA A40

Compute (FP32-eq)

1.33x

Need 1.33x A40

VRAM

0.17x

A40 has 6.00x more

AMD MI210

Compute (FP32-eq)

1.10x

Need 1.10x MI210

VRAM

0.13x

MI210 has 8.00x more

AMD MI350X

Compute (FP32-eq)

0.10x

MI350X is 10.50x faster

VRAM

0.03x

MI350X has 36.00x more

FuriosaAI Warboy

VRAM

0.25x

Warboy has 4.00x more

AMD MI300A

Compute (FP32-eq)

0.20x

MI300A is 4.90x faster

VRAM

0.06x

MI300A has 16.00x more

To match 1x FuriosaAI Warboy

NVIDIA A40

VRAM

0.67x

A40 has 1.50x more

AMD MI210

VRAM

0.50x

MI210 has 2.00x more

AMD MI350X

VRAM

0.11x

MI350X has 9.00x more

Tenstorrent Grayskull

VRAM

4.00x

Need 4.00x Grayskull

AMD MI300A

VRAM

0.25x

MI300A has 4.00x more

To match 1x AMD MI300A

NVIDIA A40

Compute (FP32-eq)

6.54x

Need 6.54x A40

FP32 Compute

3.28x

Need 3.28x A40

VRAM

2.67x

Need 2.67x A40

Memory Bandwidth

7.61x

Need 7.61x A40

AMD MI210

Compute (FP32-eq)

5.42x

Need 5.42x MI210

FP32 Compute

2.71x

Need 2.71x MI210

VRAM

2.00x

Need 2.00x MI210

Memory Bandwidth

3.24x

Need 3.24x MI210

AMD MI350X

Compute (FP32-eq)

0.47x

MI350X is 2.14x faster

VRAM

0.44x

MI350X has 2.25x more

Tenstorrent Grayskull

Compute (FP32-eq)

4.90x

Need 4.90x Grayskull

VRAM

16.00x

Need 16.00x Grayskull

FuriosaAI Warboy

VRAM

4.00x

Need 4.00x Warboy

Pricing

Price Type	A40	MI210	MI350X	Grayskull	Warboy	MI300A
CAPEX (Street Price)	—	$6,000	—	—	—	—
OPEX (per hour)	—	—	—	—	—	—
Price per TFLOPs (FP32-eq)	—	$66	—	—	—	—