AMD Radeon Instinct MI300X vs Matrox LUMA A310

Comparison of AMD Radeon Instinct MI300X with 192 GB HBM3 and 19,456 cores vs Matrox LUMA A310 with 4 GB GDDR6 and 768 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

AMD Radeon Instinct MI300X

AMD Radeon Instinct MI300X

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

Matrox LUMA A310

2.1

Matrox LUMA A310

2.1

Contents:

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

Memory

Memory Size

192 GB
🔥 4 ГБ

Memory Type

HBM3 GDDR6

Memory Bandwidth

10.3 TB/s 124.0 GB/s

Memory Bus Width

8,192 бит 64 бит

ML Performance

FP16 (Half Precision)

653.7 TFLOPS
🔥 5.376 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

81.72 TFLOPS
🔥 2.688 TFLOPS

FP64 (Double Precision)

81.72 TFLOPS
🔥 0.672 TFLOPS

CUDA Cores

19,456
🔥 768

RT Cores

No
🔥 6

Architecture & Compatibility

GPU Architecture

CDNA 3.0 Generation 12.7

SM (Streaming Multiprocessor)

No No

PCIe Version

PCIe 5.0 x16 PCIe 4.0 x8

ML Software Support

CUDA Version

No No

Clocks & Performance

Base Clock

1,000
🔥 +75% 1,750

Boost Clock

2,100
🔥 1,750

Memory Clock

2,525
🔥 1,937

Power Consumption

TDP/TGP

750 W
🔥 -96% 30 W

Recommended PSU

1150 W
🔥 -83% 200 W

Power Connector

None None

Rendering

Texture Units (TMU)

1,216
🔥 32

ROP

No
🔥 6

L2 Cache

16 MB
🔥 4 MB

Benchmarks

MLPerf, llama2-70b-99.9 (UNSET)

1 983 tokens/s

MLPerf, llama2-70b-99.9 (fp16)

1 740 tokens/s

MLPerf, llama2-70b-99.9 (fp8)

1 057 tokens/s

MLPerf, llama3.1-405b (UNSET)

30.4 tokens/s

MLPerf, llama3.1-405b (fp16)

34.8 tokens/s

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

232.9 tokens/s

MLPerf, mixtral-8x7b (fp8)

5 975 tokens/s

Additional

Slots

OAM Module Single-slot

Release Date

Dec. 6, 2023 April 27, 2023

Display Outputs

No outputs
4x mini-DisplayPort 2.0

Renting is cheaper than buying