NVIDIA H200 SXM 141 GB vs TongFang Hongkong RTX 4060 Max-Q

Comparison of NVIDIA H200 SXM 141 GB with 141 GB HBM3e and 16,896 cores vs TongFang Hongkong RTX 4060 Max-Q with 8 GB GDDR6 and 3,072 cores.

Loading...

Performance Rating

NVIDIA H200 SXM 141 GB outperforms TongFang Hongkong RTX 4060 Max-Q by 1,129.56% 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

TongFang Hongkong RTX 4060 Max-Q

5.5

TongFang Hongkong RTX 4060 Max-Q

5.5

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 GDDR6

Memory Bandwidth

🔥 4.89 TB/s
256.0 GB/s

Memory Bus Width

6,144 бит 128 бит

ML Performance

FP16 (Half Precision)

🔥 +2,863% 267.6 TFLOPS
9.032 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +641% 66.91 TFLOPS
9.032 TFLOPS

FP64 (Double Precision)

33.45 TFLOPS
🔥 +421,724% 141,100.0 TFLOPS

CUDA Cores

🔥 +450% 16,896
3,072

RT Cores

No
🔥 24

Architecture & Compatibility

GPU Architecture

Hopper Ada Lovelace

SM (Streaming Multiprocessor)

🔥 +450% 132
24

PCIe Version

PCIe 5.0 x16 PCIe 4.0 x8

ML Software Support

CUDA Version

🔥 9.0
8.9

Clocks & Performance

Base Clock

🔥 +32% 1,500
1,140

Boost Clock

🔥 +35% 1,980
1,470

Memory Clock

1,593
🔥 +26% 2,000

Power Consumption

TDP/TGP

700 W
🔥 -95% 35 W

Recommended PSU

1100 W No

Power Connector

8-pin EPS None

Rendering

Texture Units (TMU)

🔥 +450% 528
96

ROP

No
🔥 24

L2 Cache

🔥 +56% 50 MB
32 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
IGP

Release Date

Nov. 18, 2024 Jan. 3, 2023

Display Outputs

No outputs
Portable Device Dependent

Renting is cheaper than buying