NVIDIA RTX PRO 4000 Blackwell SFF vs ZOTAC RTX 3060 Thunderbolt HA

Comparison of NVIDIA RTX PRO 4000 Blackwell SFF with 24 GB GDDR7 and 8,960 cores vs ZOTAC RTX 3060 Thunderbolt HA with 12 GB GDDR6 and 3,584 cores.

Loading...

Performance Rating

NVIDIA RTX PRO 4000 Blackwell SFF outperforms ZOTAC RTX 3060 Thunderbolt HA by 69.04% 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

ZOTAC RTX 3060 Thunderbolt HA

7.9

ZOTAC RTX 3060 Thunderbolt HA

7.9

Contents:

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

Memory

Memory Size

🔥 +100% 24 ГБ
12 GB

Memory Type

GDDR7 GDDR6

Memory Bandwidth

432.0 GB/s 360.0 GB/s

Memory Bus Width

192 бит 192 бит

ML Performance

FP16 (Half Precision)

🔥 +101% 25.66 TFLOPS
12.74 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +101% 25.66 TFLOPS
12.74 TFLOPS

FP64 (Double Precision)

🔥 +102% 0.401 TFLOPS
0.199 TFLOPS

CUDA Cores

🔥 +150% 8,960
3,584

RT Cores

🔥 +150% 70
28

Architecture & Compatibility

GPU Architecture

Blackwell 2.0 Ampere

SM (Streaming Multiprocessor)

🔥 +150% 70
28

PCIe Version

PCIe 5.0 x8 PCIe 4.0 x16

ML Software Support

CUDA Version

🔥 12.0
8.6

Clocks & Performance

Base Clock

577
🔥 +129% 1,320

Boost Clock

1,432
🔥 +24% 1,777

Memory Clock

1,125
🔥 +67% 1,875

Power Consumption

Recommended PSU

🔥 -44% 250 W
450 W

Power Connector

None 1x 8-pin

TDP/TGP

🔥 -59% 70 W
170 W

Rendering

Texture Units (TMU)

🔥 +150% 280
112

ROP

🔥 +150% 70
28

L2 Cache

🔥 +1,500% 48 MB
3 MB

Additional

Slots

Dual-slot Dual-slot

Release Date

Aug. 11, 2025 Jan. 12, 2021

Display Outputs

4x mini-DisplayPort 2.1b
1x HDMI 2.1
3x DisplayPort 1.4a

Renting is cheaper than buying