AMD Radeon Instinct MI355X vs NVIDIA GeForce RTX 4090

Comparison of AMD Radeon Instinct MI355X with 288 GB HBM3e and 16,384 cores vs NVIDIA GeForce RTX 4090 with 24 GB GDDR6X and 16,384 cores.

Loading...

Performance Rating

AMD Radeon Instinct MI355X outperforms NVIDIA GeForce RTX 4090 by 191.7% 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 GeForce RTX 4090

32.1

NVIDIA GeForce RTX 4090

32.1

Contents:

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

Memory

Memory Size

🔥 +1,100% 288 ГБ
24 ГБ

Memory Type

HBM3e GDDR6X

Memory Bandwidth

🔥 +711% 8.19 TB/s
1.01 TB/s

Memory Bus Width

8,192 бит 384 бит

ML Performance

FP16 (Half Precision)

🔥 +662% 629.1 TFLOPS
82.58 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

78.64 TFLOPS
🔥 +5% 82.58 TFLOPS

FP64 (Double Precision)

🔥 78.64 TFLOPS
No

CUDA Cores

16,384 16,384

RT Cores

No
🔥 128

Architecture & Compatibility

GPU Architecture

CDNA 4.0 Ada Lovelace

SM (Streaming Multiprocessor)

No
🔥 128

PCIe Version

PCIe 5.0 x16 PCIe 4.0 x16

ML Software Support

CUDA Version

No 8.9

Clocks & Performance

Base Clock

1,000
🔥 +124% 2,235

Boost Clock

2,400
🔥 +5% 2,520

Memory Clock

🔥 +52% 2,000
1,313

Power Consumption

TDP/TGP

1400 W
🔥 -68% 450 W

Recommended PSU

1800 W
🔥 -53% 850 W

Power Connector

None 1x 16-pin

Rendering

Texture Units (TMU)

🔥 +100% 1,024
512

ROP

No
🔥 128

L2 Cache

16 MB
🔥 +350% 72 MB

Benchmarks

MLPerf, llama2-70b-99.9 (mxfp4)

10 129 tokens/s

MLPerf, llama3.1-405b (mxfp4)

252.4 tokens/s

MLPerf, llama3.1-8b-edge (fp32)

44.7 tokens/s

llama.cpp, llama 7B Q4_0

154.7 tokens/s

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

189.0 tokens/s

Geekbench AI, FP16

53 496 points

Geekbench AI, INT8

29 155 points

Geekbench AI, FP32

39 033 points

Additional

Slots

OAM Module Triple-slot

Release Date

July 15, 2025 Sept. 20, 2022

Display Outputs

No outputs
1x HDMI 2.1
3x DisplayPort 1.4a

Renting is cheaper than buying