AMD Radeon Instinct MI355X vs NVIDIA CMP 170HX 8 GB

Comparison of AMD Radeon Instinct MI355X with 288 GB HBM3e and 16,384 cores vs NVIDIA CMP 170HX 8 GB with 8 GB HBM2e and 4,480 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

AMD Radeon Instinct MI355X

AMD Radeon Instinct MI355X

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

NVIDIA CMP 170HX 8 GB

NVIDIA CMP 170HX 8 GB

Contents:

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

Memory

Memory Size

288 GB 8 GB

Memory Type

HBM3e HBM2e

Memory Bandwidth

8.19 TB/s 1.49 TB/s

Memory Bus Width

8,192 бит 4,096 бит

ML Performance

FP16 (Half Precision)

629.1 TFLOPS 50.53 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

78.64 TFLOPS 12.63 TFLOPS

FP64 (Double Precision)

78.64 TFLOPS 6.317 TFLOPS

CUDA Cores

16,384 4,480

RT Cores

No No

Architecture & Compatibility

GPU Architecture

CDNA 4.0 Ampere

SM (Streaming Multiprocessor)

No 70

PCIe Version

PCIe 5.0 x16 PCIe 1.0 x4

ML Software Support

CUDA Version

No 8.0

Clocks & Performance

Base Clock

1,000 1,140

Boost Clock

2,400 1,410

Memory Clock

2,000 1,458

Power Consumption

TDP/TGP

1400 W
🔥 -82% 250 W

Recommended PSU

1800 W
🔥 -67% 600 W

Power Connector

None 2x 8-pin

Rendering

Texture Units (TMU)

1,024 280

ROP

No No

L2 Cache

32 MB 8 MB

Benchmarks

MLPerf, llama2-70b-99.9 (mxfp4)

10 129 tokens/s

MLPerf, llama3.1-405b (mxfp4)

252.4 tokens/s

Additional

Slots

OAM Module IGP

Release Date

March 12, 2025 Sept. 1, 2021

Display Outputs

No outputs
No outputs

Renting is cheaper than buying