NVIDIA RTX PRO 6000 Blackwell Server vs PNY DGX Spark

Comparison NVIDIA RTX PRO 6000 Blackwell Server with 96 GB GDDR7 and 24,064 cores vs PNY DGX Spark with 128 GB LPDDR5X and 6,144 cores.

Loading...

Performance Rating

NVIDIA RTX PRO 6000 Blackwell Server outperforms PNY DGX Spark by 230.86% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

NVIDIA RTX PRO 6000 Blackwell Server

57.5

NVIDIA RTX PRO 6000 Blackwell Server

57.5
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

PNY DGX Spark

17.4

PNY DGX Spark

17.4

Expert Comparison

NVIDIA RTX PRO 6000 Blackwell Server значительно превосходит PNY DGX Spark в производительности и количестве ядер. RTX PRO 6000 Blackwell Server имеет 24064 ядра против 6144 у DGX Spark, 96 ГБ GDDR7 против 128 ГБ LPDDR5X памяти, а также более высокую пропускную способность (1.79 ТБ/с против 273.2 ГБ/с) и большую TDP (600 Вт против 140 Вт). RTX PRO 6000 Blackwell Server лучше подходит для вычислений на больших объемах данных, машинного обучения и графики. DGX Spark более энергоэффективен и лучше подходит для менее требовательных задач или малых систем.

Contents:

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

Memory

Memory Size

96 GB
🔥 +33% 128 ГБ

Memory Type

GDDR7 LPDDR5X

Memory Bandwidth

🔥 1.79 TB/s
273.2 GB/s

Memory Bus Width

512 бит 256 бит

ML Performance

FP16 (Half Precision)

🔥 +324% 126.0 TFLOPS
29.71 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +324% 126.0 TFLOPS
29.71 TFLOPS

FP64 (Double Precision)

🔥 +324% 1.968 TFLOPS
0.4643 TFLOPS

CUDA Cores

🔥 +292% 24,064
6,144

RT Cores

🔥 +292% 188
48

Architecture & Compatibility

GPU Architecture

Blackwell 2.0 Blackwell 2.0

SM (Streaming Multiprocessor)

🔥 +292% 188
48

PCIe Version

PCIe 5.0 x16 PCIe 5.0 x16

ML Software Support

CUDA Version

12.0
🔥 12.1

CUDA Toolkit (first supported)

v12 v12

CUDA Toolkit status

Supported Supported

Clocks & Performance

Base Clock

1,590
🔥 +5% 1,665

Boost Clock

🔥 +8% 2,617
2,418

Memory Clock

🔥 +64% 1,750
1,067

Power Consumption

Recommended PSU

1000 W
🔥 -70% 300 W

Power Connector

1x 16-pin None

TDP/TGP

600 W
🔥 -77% 140 W

Rendering

Texture Units (TMU)

🔥 +96% 752
384

ROP

🔥 +292% 188
48

L2 Cache

128 MB
🔥 50 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

Dual-slot IGP

Release Date

March 18, 2025 Oct. 15, 2025

Display Outputs

4x DisplayPort 2.1b
1x HDMI

Renting is cheaper than buying