Gainward GT 640 vs NVIDIA RTX PRO 6000 Blackwell Server

Comparison of Gainward GT 640 with 1,024 GB GDDR5 and 384 cores vs NVIDIA RTX PRO 6000 Blackwell Server with 96 GB GDDR7 and 24,064 cores.

Loading...

Performance Rating

Gainward GT 640 outperforms NVIDIA RTX PRO 6000 Blackwell Server by 50.1% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

Gainward GT 640

86.3

Gainward GT 640

86.3
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

NVIDIA RTX PRO 6000 Blackwell Server

57.5

NVIDIA RTX PRO 6000 Blackwell Server

57.5

Expert Comparison

Gainward GT 640 более экономичный вариант с низким TDP 49 Вт и подходит для базовых задач, таких как обработка видео и простые игры. NVIDIA RTX PRO 6000 Blackwell Server намного мощнее с 24064 ядрами и 1.79 TB/s пропускной способностью, что делает его идеальным для серверных задач и высокопроизводительных вычислений.

Contents:

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

Memory

Memory Size

1,024 ГБ
🔥 96 ГБ

Memory Type

GDDR5 GDDR7

Memory Bandwidth

40.06 GB/s
🔥 1.79 TB/s

Memory Bus Width

64 бит 512 бит

ML Performance

FP16 (Half Precision)

No
🔥 126.0 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

803.3 TFLOPS
🔥 126.0 TFLOPS

FP64 (Double Precision)

0.03347 TFLOPS
🔥 +5,780% 1.968 TFLOPS

CUDA Cores

384
🔥 +6,167% 24,064

RT Cores

No
🔥 188

Architecture & Compatibility

GPU Architecture

Kepler 2.0 Blackwell 2.0

SM (Streaming Multiprocessor)

No
🔥 188

PCIe Version

PCIe 2.0 x8 PCIe 5.0 x16

ML Software Support

CUDA Version

3.5
🔥 12.0

Clocks & Performance

Base Clock

No
🔥 1,590

Boost Clock

No
🔥 2,617

Memory Clock

1,252
🔥 +40% 1,750

Power Consumption

TDP/TGP

🔥 -92% 49 W
600 W

Recommended PSU

🔥 -80% 200 W
1000 W

Power Connector

None 1x 16-pin

Rendering

Texture Units (TMU)

32
🔥 +2,250% 752

ROP

No
🔥 188

L2 Cache

512 KB
🔥 128 MB

Benchmarks

MLPerf, llama2-70b-99.9 (fp4)

3 250 tokens/s

MLPerf, llama3.1-8b (fp4)

5 758 tokens/s

Geekbench AI, FP16

53 322 points

Geekbench AI, INT8

28 264 points

Geekbench AI, FP32

37 299 points

MLPerf, mixtral-8x7b (fp8)

3 767 tokens/s

Additional

Slots

Single-slot Dual-slot

Release Date

May 29, 2013 March 18, 2025

Display Outputs

1x DVI
1x HDMI 1.4a
1x VGA
4x DisplayPort 2.1b

Renting is cheaper than buying