Intel Arc Pro A40 vs NVIDIA H200 SXM 141 GB

Comparison Intel Arc Pro A40 with 6 GB GDDR6 and 1,024 cores vs NVIDIA H200 SXM 141 GB with 141 GB HBM3e and 16,896 cores.

Loading...

Performance Rating

NVIDIA H200 SXM 141 GB outperforms Intel Arc Pro A40 by 2,264.21% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

Intel Arc Pro A40

2.9

Intel Arc Pro A40

2.9
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

NVIDIA H200 SXM 141 GB

67.4

NVIDIA H200 SXM 141 GB

67.4

Expert Comparison

Intel Arc Pro A40 более энергоэффективен и дешевле, но значительно уступает NVIDIA H200 SXM по мощности и объему памяти. H200 SXM предназначена для обработки больших данных и AI-задач благодаря своим высоким показателям пропускной способности и количеству ядер.

Contents:

Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering Benchmarks Additional

Memory

Memory Size

6 GB
🔥 +2,250% 141 ГБ

Memory Type

GDDR6 HBM3e

Memory Bandwidth

192.0 GB/s
🔥 4.89 TB/s

Memory Bus Width

96 бит 6,144 бит

ML Performance

FP16 (Half Precision)

6.963 TFLOPS
🔥 +3,743% 267.6 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

3.482 TFLOPS
🔥 +1,822% 66.91 TFLOPS

FP64 (Double Precision)

0.8704 TFLOPS
🔥 +3,743% 33.45 TFLOPS

CUDA Cores

1,024
🔥 +1,550% 16,896

RT Cores

🔥 8
No

Architecture & Compatibility

GPU Architecture

Xe-HPG Hopper

SM (Streaming Multiprocessor)

No
🔥 132

PCIe Version

PCIe 4.0 x8 PCIe 5.0 x16

ML Software Support

CUDA Version

No 9.0

CUDA Toolkit (first supported)

v12

CUDA Toolkit status

Supported Supported

Clocks & Performance

Base Clock

1,500 1,500

Boost Clock

1,700
🔥 +16% 1,980

Memory Clock

🔥 +26% 2,000
1,593

Power Consumption

Recommended PSU

🔥 -77% 250 W
1100 W

Power Connector

None 8-pin EPS

TDP/TGP

🔥 -93% 50 W
700 W

Rendering

Texture Units (TMU)

64
🔥 +725% 528

ROP

🔥 8
No

L2 Cache

4 MB
🔥 +1,150% 50 MB

Benchmarks

MLPerf, llama2-70b-99.9 (UNSET)

3 534 tokens/s

MLPerf, llama2-70b-99.9 (fp16)

3 553 tokens/s

MLPerf, llama2-70b-99.9 (fp8)

2 444 tokens/s

MLPerf, llama3.1-405b (fp16)

40.8 tokens/s

MLPerf, llama3.1-405b (fp8)

25.3 tokens/s

MLPerf, llama3.1-8b (fp8)

5 161 tokens/s

MLPerf, deepseek-r1 (fp8)

1 113 tokens/s

MLPerf, mixtral-8x7b (fp8)

7 132 tokens/s

Additional

Slots

Single-slot
🔥 SXM Module

Release Date

Aug. 8, 2022 Nov. 18, 2024

Display Outputs

4x mini-DisplayPort 2.0
No outputs

Rent on Intelion Cloud

H200 (141GB)

Available for rent on Intelion Cloud

from 375.0 RUB/hour Rent
Pay-as-you-go, per-second billing
Prices include 22% VAT
EDI & closing documents
Server ready in 3 minutes
Russian-speaking support
Card, SBP or invoice