NVIDIA GRID RTX T10-8 vs NVIDIA H200 SXM 141 GB

Comparison of NVIDIA GRID RTX T10-8 with 8 GB GDDR6 and 3,584 cores vs NVIDIA H200 SXM 141 GB with 141 GB HBM3e and 16,896 cores.

Loading...

Performance Rating

NVIDIA H200 SXM 141 GB outperforms NVIDIA GRID RTX T10-8 by 758.34% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

NVIDIA GRID RTX T10-8

7.9

NVIDIA GRID RTX T10-8

7.9
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

8 ГБ
🔥 +1,662% 141 ГБ

Memory Type

GDDR6 HBM3e

Memory Bandwidth

672.0 GB/s
🔥 4.89 TB/s

Memory Bus Width

384 бит 6,144 бит

ML Performance

FP16 (Half Precision)

20.0 TFLOPS
🔥 +1,238% 267.6 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

9.999 TFLOPS
🔥 +569% 66.91 TFLOPS

FP64 (Double Precision)

0.3125 TFLOPS
🔥 +10,604% 33.45 TFLOPS

CUDA Cores

3,584
🔥 +371% 16,896

RT Cores

56 No

Architecture & Compatibility

GPU Architecture

Turing Hopper

SM (Streaming Multiprocessor)

56
🔥 +136% 132

PCIe Version

PCIe 3.0 x16 PCIe 5.0 x16

ML Software Support

CUDA Version

7.5
🔥 9.0

Clocks & Performance

Base Clock

1,065
🔥 +41% 1,500

Boost Clock

1,395
🔥 +42% 1,980

Memory Clock

1,750
🔥 1,593

Power Consumption

TDP/TGP

🔥 -79% 150 W
700 W

Recommended PSU

🔥 -59% 450 W
1100 W

Power Connector

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

Rendering

Texture Units (TMU)

224
🔥 +136% 528

ROP

56 No

L2 Cache

6 MB
🔥 +733% 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

Dual-slot
🔥 SXM Module

Release Date

No Nov. 18, 2024

Display Outputs

No outputs
No outputs

Renting is cheaper than buying