NVIDIA H200 SXM 141 GB vs NVIDIA Quadro RTX 8000 Passive

Comparison of NVIDIA H200 SXM 141 GB with 141 GB HBM3e and 16,896 cores vs NVIDIA Quadro RTX 8000 Passive with 48 GB GDDR6 and 4,608 cores.

Loading...

Performance Rating

NVIDIA H200 SXM 141 GB outperforms NVIDIA Quadro RTX 8000 Passive by 434.34% 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

NVIDIA Quadro RTX 8000 Passive

12.6

NVIDIA Quadro RTX 8000 Passive

12.6

Contents:

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

Memory

Memory Size

🔥 +194% 141 ГБ
48 ГБ

Memory Type

HBM3e GDDR6

Memory Bandwidth

🔥 4.89 TB/s
624.0 GB/s

Memory Bus Width

6,144 бит 384 бит

ML Performance

FP16 (Half Precision)

🔥 +796% 267.6 TFLOPS
29.86 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +348% 66.91 TFLOPS
14.93 TFLOPS

FP64 (Double Precision)

🔥 +7,069% 33.45 TFLOPS
0.4666 TFLOPS

CUDA Cores

🔥 +267% 16,896
4,608

RT Cores

No 72

Architecture & Compatibility

GPU Architecture

Hopper Turing

SM (Streaming Multiprocessor)

🔥 +83% 132
72

PCIe Version

PCIe 5.0 x16 PCIe 3.0 x16

ML Software Support

CUDA Version

🔥 9.0
7.5

Clocks & Performance

Base Clock

🔥 +22% 1,500
1,230

Boost Clock

🔥 +22% 1,980
1,620

Memory Clock

🔥 1,593
1,625

Power Consumption

TDP/TGP

700 W
🔥 -63% 260 W

Recommended PSU

1100 W
🔥 -45% 600 W

Power Connector

8-pin EPS 1x 6-pin + 1x 8-pin

Rendering

Texture Units (TMU)

🔥 +83% 528
288

ROP

No 72

L2 Cache

🔥 +733% 50 MB
6 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. 13, 2018

Display Outputs

No outputs
No outputs

Renting is cheaper than buying