Compare XPUs

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

Select XPUs to Compare

Clear all (3)Jump to results

Filter by Vendor

Showing 72 XPUs • 3 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 A10	SambaNova SN40L	NVIDIA GeForce RTX 5090
Architecture	Ampere	Reconfigurable Dataflow Unit (RDU)	Blackwell
Form Factor	PCIe	—	PCIe
VRAM	24 GB	640 GB	32 GB
Memory Bandwidth	600 GB/s	—	1,792 GB/s
TFLOPs (FP32)	31.2	—	104.8
TFLOPs (FP16)	125	—	209.5
TFLOPs	125	—	209.5
TFLOPs (FP8)	—	—	—
TDP	150 W	700 W	575 W
Launch Date	Apr 2021	Jan 2024	Jan 2025

Efficiency Metrics

Metric	A10	SN40L	GeForce RTX 5090
TFLOPs per Watt (FP32-eq)	0.42	—	0.18
Memory Bandwidth per GB	25.0 GB/s	—	56.0 GB/s

Performance Equivalence

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

To match 1x NVIDIA A10

SambaNova SN40L

VRAM

0.04x

SN40L has 26.67x more

NVIDIA GeForce RTX 5090

Compute (FP32-eq)

0.60x

GeForce RTX 5090 is 1.68x faster

FP32 Compute

0.30x

GeForce RTX 5090 is 3.36x faster

VRAM

0.75x

GeForce RTX 5090 has 1.33x more

Memory Bandwidth

0.33x

GeForce RTX 5090 has 2.99x more

To match 1x SambaNova SN40L

NVIDIA A10

VRAM

26.67x

Need 26.67x A10

NVIDIA GeForce RTX 5090

VRAM

20.00x

Need 20.00x GeForce RTX 5090

To match 1x NVIDIA GeForce RTX 5090

NVIDIA A10

Compute (FP32-eq)

1.68x

Need 1.68x A10

FP32 Compute

3.36x

Need 3.36x A10

VRAM

1.33x

Need 1.33x A10

Memory Bandwidth

2.99x

Need 2.99x A10

SambaNova SN40L

VRAM

0.05x

SN40L has 20.00x more

Pricing

Price Type	A10	SN40L	GeForce RTX 5090
CAPEX (Street Price)	—	—	—
OPEX (per hour)	—	—	—
Price per TFLOPs (FP32-eq)	—	—	—