NVIDIA RTX PRO 4000 Blackwell SFF vs Palit CMP 30HX

Comparison of NVIDIA RTX PRO 4000 Blackwell SFF with 24 GB GDDR7 and 8,960 cores vs Palit CMP 30HX with 6 GB GDDR6 and 1,408 cores.

Loading...

Performance Rating

NVIDIA RTX PRO 4000 Blackwell SFF outperforms Palit CMP 30HX by 217.46% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

NVIDIA RTX PRO 4000 Blackwell SFF

13.3

NVIDIA RTX PRO 4000 Blackwell SFF

13.3
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

Palit CMP 30HX

4.2

Palit CMP 30HX

4.2

Contents:

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

Memory

Memory Size

🔥 +300% 24 ГБ
6 GB

Memory Type

GDDR7 GDDR6

Memory Bandwidth

432.0 GB/s 336.0 GB/s

Memory Bus Width

192 бит 192 бит

ML Performance

FP16 (Half Precision)

🔥 +155% 25.66 TFLOPS
10.05 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +410% 25.66 TFLOPS
5.027 TFLOPS

FP64 (Double Precision)

🔥 +155% 0.401 TFLOPS
0.1571 TFLOPS

CUDA Cores

🔥 +536% 8,960
1,408

RT Cores

🔥 70
No

Architecture & Compatibility

GPU Architecture

Blackwell 2.0 Turing

SM (Streaming Multiprocessor)

🔥 +218% 70
22

PCIe Version

PCIe 5.0 x8 PCIe 1.0 x4

ML Software Support

CUDA Version

🔥 12.0
7.5

Clocks & Performance

Base Clock

577
🔥 +165% 1,530

Boost Clock

1,432
🔥 +25% 1,785

Memory Clock

1,125
🔥 +56% 1,750

Power Consumption

Recommended PSU

🔥 -17% 250 W
300 W

Power Connector

None 1x 8-pin

TDP/TGP

🔥 -44% 70 W
125 W

Rendering

Texture Units (TMU)

🔥 +218% 280
88

ROP

🔥 70
No

L2 Cache

🔥 48 MB
1536 KB

Additional

Slots

Dual-slot Dual-slot

Release Date

Aug. 11, 2025 Feb. 25, 2021

Display Outputs

4x mini-DisplayPort 2.1b
No outputs

Renting is cheaper than buying