Compare XPUs

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

Select XPUs to Compare

Clear all (5)Jump to results

Filter by Vendor

Showing 128 XPUs • 5 selected

Alibaba

Hanguang 800

AMD

MI100

23.1 TFLOPs

AMD

MI210

181 TFLOPs

AMD

MI250X

383 TFLOPs

AMD

MI300X

1,307 TFLOPs

AMD

MI325X

1,400 TFLOPs

AMD

MI350X

2,100 TFLOPs

AMD

MI355X

5,300 TFLOPs

AMD

Radeon Pro V520

23.04 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

FuriosaAI

RNGD (Renegade)

256 TFLOPs

FuriosaAI

Warboy

Google

TPU v4

275 TFLOPs

Google

TPU v5e

197 TFLOPs

Google

TPU v5p

459 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	Tenstorrent Grayskull	Enflame Technology CloudBlazer T20	Intel Data Center GPU Max 1100	NVIDIA RTX 6000 Ada Generation	AMD MI355X
Architecture	Tensix Core	GCU (AI Graphics Unit)	Ponte Vecchio	Ada Lovelace	CDNA 4
Form Factor	—	—	OAM	PCIe	OAM
VRAM	8 GB	32 GB	48 GB	48 GB	288 GB
Memory Bandwidth	—	—	1,229 GB/s	960 GB/s	8,000 GB/s
TFLOPs (FP32)	—	—	22	91.1	320
TFLOPs (FP16)	—	—	177	—	—
TFLOPs	200	—	177	182.5	5,300
TFLOPs (FP8)	—	—	—	—	—
TDP	65 W	250 W	300 W	300 W	1000 W
Launch Date	Mar 2021	Nov 2021	Jan 2023	Sep 2022	Jun 2025

Efficiency Metrics

Metric	Grayskull	CloudBlazer T20	Data Center GPU Max 1100	RTX 6000 Ada Generation	MI355X
TFLOPs per Watt (FP32-eq)	1.54	—	0.29	0.30	2.65
Memory Bandwidth per GB	—	—	25.6 GB/s	20.0 GB/s	27.8 GB/s

Performance Equivalence

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

To match 1x Tenstorrent Grayskull

Enflame Technology CloudBlazer T20

VRAM

0.25x

CloudBlazer T20 has 4.00x more

Intel Data Center GPU Max 1100

Compute (FP32-eq)

1.13x

Need 1.13x Data Center GPU Max 1100

VRAM

0.17x

Data Center GPU Max 1100 has 6.00x more

NVIDIA RTX 6000 Ada Generation

Compute (FP32-eq)

1.10x

Need 1.10x RTX 6000 Ada Generation

VRAM

0.17x

RTX 6000 Ada Generation has 6.00x more

AMD MI355X

Compute (FP32-eq)

0.04x

MI355X is 26.50x faster

VRAM

0.03x

MI355X has 36.00x more

To match 1x Enflame Technology CloudBlazer T20

Tenstorrent Grayskull

VRAM

4.00x

Need 4.00x Grayskull

Intel Data Center GPU Max 1100

VRAM

0.67x

Data Center GPU Max 1100 has 1.50x more

NVIDIA RTX 6000 Ada Generation

VRAM

0.67x

RTX 6000 Ada Generation has 1.50x more

AMD MI355X

VRAM

0.11x

MI355X has 9.00x more

To match 1x Intel Data Center GPU Max 1100

Tenstorrent Grayskull

Compute (FP32-eq)

0.89x

Grayskull is 1.13x faster

VRAM

6.00x

Need 6.00x Grayskull

Enflame Technology CloudBlazer T20

VRAM

1.50x

Need 1.50x CloudBlazer T20

NVIDIA RTX 6000 Ada Generation

Compute (FP32-eq)

0.97x

RTX 6000 Ada Generation is 1.03x faster

FP32 Compute

0.24x

RTX 6000 Ada Generation is 4.14x faster

VRAM

1.00x

RTX 6000 Ada Generation has 1.00x more

Memory Bandwidth

1.28x

Need 1.28x RTX 6000 Ada Generation

AMD MI355X

Compute (FP32-eq)

0.03x

MI355X is 29.94x faster

FP32 Compute

0.07x

MI355X is 14.55x faster

VRAM

0.17x

MI355X has 6.00x more

Memory Bandwidth

0.15x

MI355X has 6.51x more

To match 1x NVIDIA RTX 6000 Ada Generation

Tenstorrent Grayskull

Compute (FP32-eq)

0.91x

Grayskull is 1.10x faster

VRAM

6.00x

Need 6.00x Grayskull

Enflame Technology CloudBlazer T20

VRAM

1.50x

Need 1.50x CloudBlazer T20

Intel Data Center GPU Max 1100

Compute (FP32-eq)

1.03x

Need 1.03x Data Center GPU Max 1100

FP32 Compute

4.14x

Need 4.14x Data Center GPU Max 1100

VRAM

1.00x

Data Center GPU Max 1100 has 1.00x more

Memory Bandwidth

0.78x

Data Center GPU Max 1100 has 1.28x more

AMD MI355X

Compute (FP32-eq)

0.03x

MI355X is 29.04x faster

FP32 Compute

0.28x

MI355X is 3.51x faster

VRAM

0.17x

MI355X has 6.00x more

Memory Bandwidth

0.12x

MI355X has 8.33x more

To match 1x AMD MI355X

Tenstorrent Grayskull

Compute (FP32-eq)

26.50x

Need 26.50x Grayskull

VRAM

36.00x

Need 36.00x Grayskull

Enflame Technology CloudBlazer T20

VRAM

9.00x

Need 9.00x CloudBlazer T20

Intel Data Center GPU Max 1100

Compute (FP32-eq)

29.94x

Need 29.94x Data Center GPU Max 1100

FP32 Compute

14.55x

Need 14.55x Data Center GPU Max 1100

VRAM

6.00x

Need 6.00x Data Center GPU Max 1100

Memory Bandwidth

6.51x

Need 6.51x Data Center GPU Max 1100

NVIDIA RTX 6000 Ada Generation

Compute (FP32-eq)

29.04x

Need 29.04x RTX 6000 Ada Generation

FP32 Compute

3.51x

Need 3.51x RTX 6000 Ada Generation

VRAM

6.00x

Need 6.00x RTX 6000 Ada Generation

Memory Bandwidth

8.33x

Need 8.33x RTX 6000 Ada Generation

Pricing

Price Type	Grayskull	CloudBlazer T20	Data Center GPU Max 1100	RTX 6000 Ada Generation	MI355X
CAPEX (Street Price)	—	—	$5,000	—	—
OPEX (per hour)	—	—	—	$0.33/hr	$20.72/hr
Price per TFLOPs (FP32-eq)	—	—	$56	—	—