NVIDIA RTX PRO 6000 Blackwell Server vs Sapphire PULSE RX 550

Comparison of NVIDIA RTX PRO 6000 Blackwell Server with 96 GB GDDR7 and 24,064 cores vs Sapphire PULSE RX 550 with 2 GB GDDR5 and 512 cores.

Loading...

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

Sapphire PULSE RX 550

0.0

Sapphire PULSE RX 550

0.0

Contents:

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

Memory

Memory Size

🔥 +4,700% 96 ГБ
2 ГБ

Memory Type

GDDR7 GDDR5

Memory Bandwidth

🔥 1.79 TB/s
112.0 GB/s

Memory Bus Width

512 бит 128 бит

ML Performance

FP16 (Half Precision)

🔥 126.0 TFLOPS
No

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 126.0 TFLOPS
No

FP64 (Double Precision)

🔥 +2,450% 1.968 TFLOPS
0.07718000000000001 TFLOPS

CUDA Cores

🔥 +4,600% 24,064
512

RT Cores

🔥 188
No

Architecture & Compatibility

GPU Architecture

Blackwell 2.0 GCN 4.0

SM (Streaming Multiprocessor)

🔥 188
No

PCIe Version

PCIe 5.0 x16 PCIe 3.0 x8

ML Software Support

CUDA Version

12.0 No

Clocks & Performance

Base Clock

🔥 +45% 1,590
1,100

Boost Clock

🔥 +117% 2,617
1,206

Memory Clock

🔥 1,750
1,750

Power Consumption

TDP/TGP

600 W
🔥 -92% 50 W

Recommended PSU

1000 W
🔥 -75% 250 W

Power Connector

1x 16-pin None

Rendering

Texture Units (TMU)

🔥 +2,250% 752
32

ROP

🔥 188
No

L2 Cache

🔥 128 MB
512 KB

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 Single-slot

Release Date

March 18, 2025 April 20, 2017

Display Outputs

4x DisplayPort 2.1b
1x DVI
1x HDMI 2.0b
1x DisplayPort 1.4a

Renting is cheaper than buying