NVIDIA RTX PRO 4000 Blackwell SFF vs NVIDIA Tesla K20m

Comparison of NVIDIA RTX PRO 4000 Blackwell SFF with 24 GB GDDR7 and 8,960 cores vs NVIDIA Tesla K20m with 5 GB GDDR5 and 2,496 cores.

Loading...

Performance Rating

NVIDIA RTX PRO 4000 Blackwell SFF outperforms NVIDIA Tesla K20m by 314.69% 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

NVIDIA Tesla K20m

3.2

NVIDIA Tesla K20m

3.2

Contents:

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

Memory

Memory Size

🔥 +380% 24 ГБ
5 GB

Memory Type

GDDR7 GDDR5

Memory Bandwidth

432.0 GB/s 208.0 GB/s

Memory Bus Width

192 бит 320 бит

ML Performance

FP16 (Half Precision)

🔥 25.66 TFLOPS
No

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +628% 25.66 TFLOPS
3.524 TFLOPS

FP64 (Double Precision)

🔥 0.401 TFLOPS
1.1748 TFLOPS

CUDA Cores

🔥 +259% 8,960
2,496

RT Cores

🔥 70
No

Architecture & Compatibility

GPU Architecture

Blackwell 2.0 Kepler

SM (Streaming Multiprocessor)

🔥 70
No

PCIe Version

PCIe 5.0 x8 PCIe 2.0 x16

ML Software Support

CUDA Version

🔥 12.0
3.5

Clocks & Performance

Base Clock

🔥 577
No

Boost Clock

🔥 1,432
No

Memory Clock

🔥 1,125
1,300

Power Consumption

Recommended PSU

🔥 -55% 250 W
550 W

Power Connector

None 1x 6-pin + 1x 8-pin

TDP/TGP

🔥 -69% 70 W
225 W

Rendering

Texture Units (TMU)

🔥 +35% 280
208

ROP

🔥 70
No

L2 Cache

🔥 48 MB
1280 KB

Additional

Slots

Dual-slot Dual-slot

Release Date

Aug. 11, 2025 Jan. 5, 2013

Display Outputs

4x mini-DisplayPort 2.1b
No outputs

Renting is cheaper than buying