AMD Radeon Instinct MI355X vs NVIDIA CMP 30HX

Comparison of AMD Radeon Instinct MI355X with 288 GB HBM3e and 16,384 cores vs NVIDIA CMP 30HX with 6 GB GDDR6 and 1,408 cores.

Loading...

Performance Rating

AMD Radeon Instinct MI355X outperforms NVIDIA CMP 30HX by 2,137.32% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

AMD Radeon Instinct MI355X

93.5

AMD Radeon Instinct MI355X

93.5
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

NVIDIA CMP 30HX

4.2

NVIDIA CMP 30HX

4.2

Contents:

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

Memory

Memory Size

🔥 +4,700% 288 ГБ
6 ГБ

Memory Type

HBM3e GDDR6

Memory Bandwidth

🔥 8.19 TB/s
336.0 GB/s

Memory Bus Width

8,192 бит 192 бит

ML Performance

FP16 (Half Precision)

🔥 +6,160% 629.1 TFLOPS
10.05 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +1,464% 78.64 TFLOPS
5.027 TFLOPS

FP64 (Double Precision)

🔥 +49,957% 78.64 TFLOPS
0.1571 TFLOPS

CUDA Cores

🔥 +1,064% 16,384
1,408

RT Cores

No No

Architecture & Compatibility

GPU Architecture

CDNA 4.0 Turing

SM (Streaming Multiprocessor)

No
🔥 22

PCIe Version

PCIe 5.0 x16 PCIe 1.0 x4

ML Software Support

CUDA Version

No 7.5

Clocks & Performance

Base Clock

1,000
🔥 +53% 1,530

Boost Clock

🔥 +34% 2,400
1,785

Memory Clock

🔥 +14% 2,000
1,750

Power Consumption

TDP/TGP

1400 W
🔥 -91% 125 W

Recommended PSU

1800 W
🔥 -83% 300 W

Power Connector

None 1x 8-pin

Rendering

Texture Units (TMU)

🔥 +1,064% 1,024
88

ROP

No No

L2 Cache

🔥 16 MB
1536 KB

Benchmarks

MLPerf, llama2-70b-99.9 (mxfp4)

10 129 tokens/s

MLPerf, llama3.1-405b (mxfp4)

252.4 tokens/s

Additional

Slots

OAM Module Dual-slot

Release Date

July 15, 2025 Feb. 25, 2021

Display Outputs

No outputs
No outputs

Renting is cheaper than buying