NVIDIA RTX PRO 6000 Blackwell Server vs VTX3D R7 250 X-Edition

Comparison of NVIDIA RTX PRO 6000 Blackwell Server with 96 GB GDDR7 and 24,064 cores vs VTX3D R7 250 X-Edition with 1,024 GB GDDR5 and 384 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

NVIDIA RTX PRO 6000 Blackwell Server

NVIDIA RTX PRO 6000 Blackwell Server

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

VTX3D R7 250 X-Edition

VTX3D R7 250 X-Edition

Contents:

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

Memory

Memory Size

96 GB 1,024 MB

Memory Type

GDDR7 GDDR5

Memory Bandwidth

1.79 TB/s 73.60 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 0.8294 TFLOPS

FP64 (Double Precision)

1.968 TFLOPS 0.0518 TFLOPS

CUDA Cores

24,064 384

RT Cores

188 No

Architecture & Compatibility

GPU Architecture

Blackwell 2.0 GCN 1.0

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

1,590 1,030

Boost Clock

2,617 1,080

Memory Clock

1,750 1,150

Power Consumption

TDP/TGP

600 W
🔥 -89% 65 W

Recommended PSU

1000 W
🔥 -75% 250 W

Power Connector

1x 16-pin None

Rendering

Texture Units (TMU)

752 24

ROP

188 No

L2 Cache

128 MB 256 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 Oct. 8, 2013

Display Outputs

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

Renting is cheaper than buying