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 MI100	Meta MTIA v1	NVIDIA GB200 Superchip	NVIDIA GeForce RTX 4070
Architecture	CDNA	Meta Training & Inference Accelerator	Blackwell	Ada Lovelace
Form Factor	PCIe	—	SXM	PCIe
VRAM	32 GB	128 GB	384 GB	12 GB
Memory Bandwidth	1,228 GB/s	—	8,000 GB/s	504 GB/s
TFLOPs (FP32)	23.1	—	90	29.1
TFLOPs (FP16)	184.6	—	5,000	58.2
TFLOPs	23.1	—	5,000	58.2
TFLOPs (FP8)	—	—	—	—
TDP	300 W	400 W	2700 W	200 W
Launch Date	Nov 2020	May 2023	Mar 2024	Apr 2023

Efficiency Metrics

Metric	MI100	MTIA v1	GB200 Superchip	GeForce RTX 4070
TFLOPs per Watt (FP32-eq)	0.08	—	0.93	0.15
Memory Bandwidth per GB	38.4 GB/s	—	20.8 GB/s	42.0 GB/s

Performance Equivalence

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

To match 1x AMD MI100

Meta MTIA v1

VRAM

0.25x

MTIA v1 has 4.00x more

NVIDIA GB200 Superchip

Compute (FP32-eq)

0.01x

GB200 Superchip is 108.23x faster

FP32 Compute

0.26x

GB200 Superchip is 3.90x faster

VRAM

0.08x

GB200 Superchip has 12.00x more

Memory Bandwidth

0.15x

GB200 Superchip has 6.51x more

NVIDIA GeForce RTX 4070

Compute (FP32-eq)

0.79x

GeForce RTX 4070 is 1.26x faster

FP32 Compute

0.79x

GeForce RTX 4070 is 1.26x faster

VRAM

2.67x

Need 2.67x GeForce RTX 4070

Memory Bandwidth

2.44x

Need 2.44x GeForce RTX 4070

To match 1x Meta MTIA v1

AMD MI100

VRAM

4.00x

Need 4.00x MI100

NVIDIA GB200 Superchip

VRAM

0.33x

GB200 Superchip has 3.00x more

NVIDIA GeForce RTX 4070

VRAM

10.67x

Need 10.67x GeForce RTX 4070

To match 1x NVIDIA GB200 Superchip

AMD MI100

Compute (FP32-eq)

108.23x

Need 108.23x MI100

FP32 Compute

3.90x

Need 3.90x MI100

VRAM

12.00x

Need 12.00x MI100

Memory Bandwidth

6.51x

Need 6.51x MI100

Meta MTIA v1

VRAM

3.00x

Need 3.00x MTIA v1

NVIDIA GeForce RTX 4070

Compute (FP32-eq)

85.91x

Need 85.91x GeForce RTX 4070

FP32 Compute

3.09x

Need 3.09x GeForce RTX 4070

VRAM

32.00x

Need 32.00x GeForce RTX 4070

Memory Bandwidth

15.87x

Need 15.87x GeForce RTX 4070

To match 1x NVIDIA GeForce RTX 4070

AMD MI100

Compute (FP32-eq)

1.26x

Need 1.26x MI100

FP32 Compute

1.26x

Need 1.26x MI100

VRAM

0.38x

MI100 has 2.67x more

Memory Bandwidth

0.41x

MI100 has 2.44x more

Meta MTIA v1

VRAM

0.09x

MTIA v1 has 10.67x more

NVIDIA GB200 Superchip

Compute (FP32-eq)

0.01x

GB200 Superchip is 85.91x faster

FP32 Compute

0.32x

GB200 Superchip is 3.09x faster

VRAM

0.03x

GB200 Superchip has 32.00x more

Memory Bandwidth

0.06x

GB200 Superchip has 15.87x more

Pricing

Price Type	MI100	MTIA v1	GB200 Superchip	GeForce RTX 4070
CAPEX (Street Price)	—	—	—	—
OPEX (per hour)	—	—	—	—
Price per TFLOPs (FP32-eq)	—	—	—	—