Intel Arc Graphics 24EU vs Intel Data Center GPU Max NEXT

Comparison of Intel Arc Graphics 24EU and 192 cores vs Intel Data Center GPU Max NEXT with 128 GB HBM2e and 20,480 cores.

Loading...

Performance Rating

A100 A100
H200 H200
MI325X MI325X

Intel Arc Graphics 24EU

0.0

Intel Arc Graphics 24EU

0.0
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

Intel Data Center GPU Max NEXT

62.1

Intel Data Center GPU Max NEXT

62.1

Contents:

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

Memory

Memory Size

No 128 ГБ

Memory Type

System Shared HBM2e

Memory Bandwidth

System Dependent
🔥 3.21 TB/s

Memory Bus Width

No 8,192 бит

ML Performance

FP16 (Half Precision)

🔥 1.536 TFLOPS
65.54 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +1,072% 768.0 TFLOPS
65.54 TFLOPS

FP64 (Double Precision)

No 65.54 TFLOPS

CUDA Cores

🔥 192
20,480

RT Cores

No 160

Architecture & Compatibility

GPU Architecture

Xe2-LPG Generation 12.5

SM (Streaming Multiprocessor)

No No

PCIe Version

Ring Bus PCIe 5.0 x16

ML Software Support

CUDA Version

No No

Clocks & Performance

Base Clock

🔥 300
900

Boost Clock

🔥 +25% 2,000
1,600

Memory Clock

No 1,565

Power Consumption

TDP/TGP

🔥 -92% 65 W
800 W

Recommended PSU

No 1200 W

Power Connector

No 8-pin EPS

Rendering

Texture Units (TMU)

🔥 12
1,280

ROP

No 160

L2 Cache

No 408 MB

Benchmarks

MLPerf, llama2-70b-99.9 (UINT4)

527.5 tokens/s

MLPerf, llama3.1-8b (UINT4)

1 337 tokens/s

MLPerf, llama3.1-8b-edge (UINT4)

1 652 tokens/s

llama.cpp, llama-2-7b-Q4_0

8.21 tokens/s

Additional

Slots

IGP OAM Module

Release Date

Oct. 24, 2024 No

Display Outputs

Motherboard Dependent
No outputs

Renting is cheaper than buying