AMD Radeon Instinct MI355X vs NVIDIA H200 NVL

Comparison of AMD Radeon Instinct MI355X with 288 GB HBM3e and 16,384 cores vs NVIDIA H200 NVL with 141 GB HBM3e and 16,896 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 H200 NVL

63.6

NVIDIA H200 NVL

63.6

Contents:

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

Memory

Memory Size

288 GB
🔥 141 ГБ

Memory Type

HBM3e HBM3e

Memory Bandwidth

8.19 TB/s
🔥 4.89 TB/s

Memory Bus Width

8,192 бит 6,144 бит

ML Performance

FP16 (Half Precision)

629.1 TFLOPS
🔥 241.3 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

78.64 TFLOPS
🔥 60.32 TFLOPS

FP64 (Double Precision)

78.64 TFLOPS
🔥 30.16 TFLOPS

CUDA Cores

16,384
🔥 +3% 16,896

RT Cores

No No

Architecture & Compatibility

GPU Architecture

CDNA 4.0 Hopper

SM (Streaming Multiprocessor)

No
🔥 132

PCIe Version

PCIe 5.0 x16 PCIe 5.0 x16

ML Software Support

CUDA Version

No 9.0

Clocks & Performance

Base Clock

1,000
🔥 +36% 1,365

Boost Clock

2,400
🔥 1,785

Memory Clock

2,000
🔥 1,593

Power Consumption

TDP/TGP

1400 W
🔥 -57% 600 W

Recommended PSU

1800 W
🔥 -44% 1000 W

Power Connector

None 8-pin EPS

Rendering

Texture Units (TMU)

1,024
🔥 528

ROP

No No

L2 Cache

32 MB
🔥 +56% 50 MB

Benchmarks

MLPerf, llama2-70b-99.9 (fp4)

7 436 tokens/s

MLPerf, llama2-70b-99.9 (mxfp4)

10 129 tokens/s

MLPerf, llama3.1-405b (fp4)

130.5 tokens/s

MLPerf, llama3.1-405b (mxfp4)

252.4 tokens/s

MLPerf, deepseek-r1 (fp4)

2 327 tokens/s

Additional

Slots

OAM Module Dual-slot

Release Date

March 12, 2025 Nov. 18, 2024

Display Outputs

No outputs
No outputs

Renting is cheaper than buying