NVIDIA RTX PRO 6000 Blackwell Max-Q vs SPARKLE Arc A310 OmniView

Comparison of NVIDIA RTX PRO 6000 Blackwell Max-Q with 96 GB GDDR7 and 24,064 cores vs SPARKLE Arc A310 OmniView with 4 GB GDDR6 and 768 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

NVIDIA RTX PRO 6000 Blackwell Max-Q

NVIDIA RTX PRO 6000 Blackwell Max-Q

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

SPARKLE Arc A310 OmniView

SPARKLE Arc A310 OmniView

Contents:

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

Memory

Memory Size

96 GB 4 GB

Memory Type

GDDR7 GDDR6

Memory Bandwidth

1.79 TB/s 124.0 GB/s

Memory Bus Width

512 бит 64 бит

ML Performance

FP16 (Half Precision)

109.7 TFLOPS 6.144 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

109.7 TFLOPS 3.072 TFLOPS

FP64 (Double Precision)

1.715 TFLOPS 0.768 TFLOPS

CUDA Cores

24,064 768

RT Cores

188 6

Architecture & Compatibility

GPU Architecture

Blackwell 2.0 Xe-HPG

SM (Streaming Multiprocessor)

188 No

PCIe Version

PCIe 5.0 x16 PCIe 4.0 x8

ML Software Support

CUDA Version

12.0 No

Clocks & Performance

Base Clock

1,035 2,000

Boost Clock

2,280 2,000

Memory Clock

1,750 1,937

Power Consumption

TDP/TGP

300 W
🔥 -75% 75 W

Recommended PSU

700 W
🔥 -64% 250 W

Power Connector

1x 16-pin None

Rendering

Texture Units (TMU)

752 32

ROP

188 6

L2 Cache

128 MB 4 MB

Benchmarks

llama.cpp, llama 7B Q4_0

270.0 tokens/s

Geekbench AI, FP16

58 178 points

Geekbench AI, INT8

29 188 points

Geekbench AI, FP32

39 339 points

Additional

Slots

Dual-slot Single-slot

Release Date

March 18, 2025 Nov. 18, 2024

Display Outputs

4x DisplayPort 2.1b
4x mini-DisplayPort 2.0
4x HDMI 2.1

Renting is cheaper than buying