NVIDIA RTX PRO 6000 Blackwell Server vs NVIDIA Tesla P100 SXM2

Comparison of NVIDIA RTX PRO 6000 Blackwell Server with 96 GB GDDR7 and 24,064 cores vs NVIDIA Tesla P100 SXM2 with 16 GB HBM2 and 3,584 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

NVIDIA RTX PRO 6000 Blackwell Server

NVIDIA RTX PRO 6000 Blackwell Server

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

NVIDIA Tesla P100 SXM2

NVIDIA Tesla P100 SXM2

Contents:

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

Memory

Memory Size

96 GB 16 GB

Memory Type

GDDR7 HBM2

Memory Bandwidth

1.79 TB/s 732.2 GB/s

Memory Bus Width

512 бит 4,096 бит

ML Performance

FP16 (Half Precision)

126.0 TFLOPS 21.22 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

126.0 TFLOPS 10.61 TFLOPS

FP64 (Double Precision)

1.968 TFLOPS 5.304 TFLOPS

CUDA Cores

24,064 3,584

RT Cores

188 No

Architecture & Compatibility

GPU Architecture

Blackwell 2.0 Pascal

SM (Streaming Multiprocessor)

188 56

PCIe Version

PCIe 5.0 x16 PCIe 3.0 x16

ML Software Support

CUDA Version

🔥 12.0
6.0

Clocks & Performance

Base Clock

1,590 1,328

Boost Clock

2,617 1,480

Memory Clock

1,750 715

Power Consumption

TDP/TGP

600 W
🔥 -50% 300 W

Recommended PSU

1000 W
🔥 -30% 700 W

Power Connector

1x 16-pin None

Rendering

Texture Units (TMU)

752 224

ROP

188 No

L2 Cache

128 MB 4 MB

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 SXM Module

Release Date

March 18, 2025 April 5, 2016

Display Outputs

4x DisplayPort 2.1b
No outputs

Renting is cheaper than buying