NVIDIA H200 SXM 141 GB vs Sapphire NITRO+ RX Vega 56 Limited Edition

Comparison of NVIDIA H200 SXM 141 GB with 141 GB HBM3e and 16,896 cores vs Sapphire NITRO+ RX Vega 56 Limited Edition with 8 GB HBM2 and 3,584 cores.

Loading...

Performance Rating

NVIDIA H200 SXM 141 GB outperforms Sapphire NITRO+ RX Vega 56 Limited Edition by 651.17% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

NVIDIA H200 SXM 141 GB

67.4

NVIDIA H200 SXM 141 GB

67.4
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

Sapphire NITRO+ RX Vega 56 Limited Edition

9.0

Sapphire NITRO+ RX Vega 56 Limited Edition

9.0

Contents:

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

Memory

Memory Size

🔥 +1,662% 141 ГБ
8 ГБ

Memory Type

HBM3e HBM2

Memory Bandwidth

🔥 4.89 TB/s
409.6 GB/s

Memory Bus Width

6,144 бит 2,048 бит

ML Performance

FP16 (Half Precision)

🔥 +1,087% 267.6 TFLOPS
22.54 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +494% 66.91 TFLOPS
11.27 TFLOPS

FP64 (Double Precision)

🔥 +4,649% 33.45 TFLOPS
0.7042999999999999 TFLOPS

CUDA Cores

🔥 +371% 16,896
3,584

RT Cores

No No

Architecture & Compatibility

GPU Architecture

Hopper GCN 5.0

SM (Streaming Multiprocessor)

🔥 132
No

PCIe Version

PCIe 5.0 x16 PCIe 3.0 x16

ML Software Support

CUDA Version

9.0 No

Clocks & Performance

Base Clock

🔥 +15% 1,500
1,305

Boost Clock

🔥 +26% 1,980
1,572

Memory Clock

🔥 +99% 1,593
800

Power Consumption

TDP/TGP

700 W
🔥 -70% 210 W

Recommended PSU

1100 W
🔥 -50% 550 W

Power Connector

8-pin EPS 3x 8-pin

Rendering

Texture Units (TMU)

🔥 +136% 528
224

ROP

No No

L2 Cache

🔥 +1,150% 50 MB
4 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

🔥 SXM Module
Dual-slot

Release Date

Nov. 18, 2024 Aug. 14, 2017

Display Outputs

No outputs
1x HDMI 2.0b
3x DisplayPort 1.4a

2x HDMI 2.0b
2x DisplayPort 1.4a

Renting is cheaper than buying