AMD Radeon Instinct MI355X vs ASUS RTX 3090 EKWB

Comparison of AMD Radeon Instinct MI355X with 288 GB HBM3e and 16,384 cores vs ASUS RTX 3090 EKWB with 24 GB GDDR6X and 10,496 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

AMD Radeon Instinct MI355X

93.5

AMD Radeon Instinct MI355X

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

ASUS RTX 3090 EKWB

ASUS RTX 3090 EKWB

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 GB

Memory Type

HBM3e GDDR6X

Memory Bandwidth

🔥 8.19 TB/s
936.2 GB/s

Memory Bus Width

8,192 бит 384 бит

ML Performance

FP16 (Half Precision)

🔥 +1,668% 629.1 TFLOPS
35.58 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +121% 78.64 TFLOPS
35.58 TFLOPS

FP64 (Double Precision)

🔥 +14,044% 78.64 TFLOPS
0.556 TFLOPS

CUDA Cores

🔥 +56% 16,384
10,496

RT Cores

No 82

Architecture & Compatibility

GPU Architecture

CDNA 4.0 Ampere

SM (Streaming Multiprocessor)

No 82

PCIe Version

PCIe 5.0 x16 PCIe 4.0 x16

ML Software Support

CUDA Version

No 8.6

Clocks & Performance

Base Clock

🔥 1,000
1,395

Boost Clock

🔥 +42% 2,400
1,695

Memory Clock

🔥 +64% 2,000
1,219

Power Consumption

TDP/TGP

1400 W
🔥 -75% 350 W

Recommended PSU

1800 W
🔥 -58% 750 W

Power Connector

None 2x 8-pin

Rendering

Texture Units (TMU)

🔥 +212% 1,024
328

ROP

No 82

L2 Cache

🔥 +167% 16 MB
6 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 Single-slot

Release Date

July 15, 2025 Sept. 1, 2020

Display Outputs

No outputs
1x HDMI 2.1
3x DisplayPort 1.4a

Renting is cheaper than buying