AMD Radeon Pro W6800X Duo vs NVIDIA H200 SXM 141 GB

Comparison of AMD Radeon Pro W6800X Duo with 32 GB GDDR6 and 3,840 cores vs NVIDIA H200 SXM 141 GB with 141 GB HBM3e and 16,896 cores.

Loading...

Performance Rating

NVIDIA H200 SXM 141 GB outperforms AMD Radeon Pro W6800X Duo by 510.88% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

AMD Radeon Pro W6800X Duo

11.0

AMD Radeon Pro W6800X Duo

11.0
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

NVIDIA H200 SXM 141 GB

67.4

NVIDIA H200 SXM 141 GB

67.4

Contents:

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

Memory

Memory Size

32 ГБ
🔥 +341% 141 ГБ

Memory Type

GDDR6 HBM3e

Memory Bandwidth

512.0 GB/s
🔥 4.89 TB/s

Memory Bus Width

256 бит 6,144 бит

ML Performance

FP16 (Half Precision)

30.21 TFLOPS
🔥 +786% 267.6 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

15.11 TFLOPS
🔥 +343% 66.91 TFLOPS

FP64 (Double Precision)

0.9442 TFLOPS
🔥 +3,443% 33.45 TFLOPS

CUDA Cores

3,840
🔥 +340% 16,896

RT Cores

🔥 60
No

Architecture & Compatibility

GPU Architecture

RDNA 2.0 Hopper

SM (Streaming Multiprocessor)

No
🔥 132

PCIe Version

Apple MPX PCIe 5.0 x16

ML Software Support

CUDA Version

No 9.0

Clocks & Performance

Base Clock

🔥 +20% 1,800
1,500

Boost Clock

1,967
🔥 +1% 1,980

Memory Clock

🔥 +26% 2,000
1,593

Power Consumption

TDP/TGP

🔥 -43% 400 W
700 W

Recommended PSU

🔥 -27% 800 W
1100 W

Power Connector

No 8-pin EPS

Rendering

Texture Units (TMU)

240
🔥 +120% 528

ROP

🔥 60
No

L2 Cache

4 MB
🔥 +1,150% 50 MB

Benchmarks

MLPerf, llama2-70b-99.9 (UNSET)

3 534 tokens/s

MLPerf, llama2-70b-99.9 (fp16)

3 553 tokens/s

MLPerf, llama2-70b-99.9 (fp8)

2 444 tokens/s

MLPerf, llama3.1-405b (fp16)

40.8 tokens/s

MLPerf, llama3.1-405b (fp8)

25.3 tokens/s

MLPerf, llama3.1-8b (fp8)

5 161 tokens/s

MLPerf, deepseek-r1 (fp8)

1 113 tokens/s

MLPerf, mixtral-8x7b (fp8)

7 132 tokens/s

Additional

Slots

Quad-slot
🔥 SXM Module

Release Date

Aug. 3, 2021 Nov. 18, 2024

Display Outputs

1x HDMI 2.1
4x Thunderbolt
No outputs

Renting is cheaper than buying