NVIDIA A2 PCIe vs NVIDIA RTX PRO 4000 Blackwell SFF

Comparison of NVIDIA A2 PCIe with 16 GB GDDR6 and 1,280 cores vs NVIDIA RTX PRO 4000 Blackwell SFF with 24 GB GDDR7 and 8,960 cores.

Loading...

Performance Rating

NVIDIA RTX PRO 4000 Blackwell SFF outperforms NVIDIA A2 PCIe by 235.1% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

NVIDIA A2 PCIe

4.0

NVIDIA A2 PCIe

4.0
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

NVIDIA RTX PRO 4000 Blackwell SFF

13.3

NVIDIA RTX PRO 4000 Blackwell SFF

13.3

Contents:

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

Memory

Memory Size

16 GB
🔥 +50% 24 ГБ

Memory Type

GDDR6 GDDR7

Memory Bandwidth

200.1 GB/s 432.0 GB/s

Memory Bus Width

128 бит 192 бит

ML Performance

FP16 (Half Precision)

4.531 TFLOPS
🔥 +466% 25.66 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

4.531 TFLOPS
🔥 +466% 25.66 TFLOPS

FP64 (Double Precision)

0.1416 TFLOPS
🔥 +183% 0.401 TFLOPS

CUDA Cores

1,280
🔥 +600% 8,960

RT Cores

10
🔥 +600% 70

Architecture & Compatibility

GPU Architecture

Ampere Blackwell 2.0

SM (Streaming Multiprocessor)

10
🔥 +600% 70

PCIe Version

PCIe 4.0 x8 PCIe 5.0 x8

ML Software Support

CUDA Version

8.6
🔥 12.0

Clocks & Performance

Base Clock

🔥 +150% 1,440
577

Boost Clock

🔥 +24% 1,770
1,432

Memory Clock

🔥 +39% 1,563
1,125

Power Consumption

Recommended PSU

250 W 250 W

Power Connector

None None

TDP/TGP

🔥 -14% 60 W
70 W

Rendering

Texture Units (TMU)

40
🔥 +600% 280

ROP

10
🔥 +600% 70

L2 Cache

2 MB
🔥 +2,300% 48 MB

Additional

Slots

Single-slot Dual-slot

Release Date

Nov. 10, 2021 Aug. 11, 2025

Display Outputs

No outputs
4x mini-DisplayPort 2.1b

Renting is cheaper than buying