NVIDIA RTX PRO 6000 Blackwell Server vs ONIX Odyssey Arc B580 OC

Comparison of NVIDIA RTX PRO 6000 Blackwell Server with 96 GB GDDR7 and 24,064 cores vs ONIX Odyssey Arc B580 OC with 12 GB GDDR6 and 2,560 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
MI250 MI250
Instinct MI300X Instinct MI300X

ONIX Odyssey Arc B580 OC

8.7

ONIX Odyssey Arc B580 OC

8.7

Contents:

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

Memory

Memory Size

96 GB
🔥 12 ГБ

Memory Type

GDDR7 GDDR6

Memory Bandwidth

1.79 TB/s 456.0 GB/s

Memory Bus Width

512 бит 192 бит

ML Performance

FP16 (Half Precision)

126.0 TFLOPS
🔥 28.06 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

126.0 TFLOPS
🔥 14.03 TFLOPS

FP64 (Double Precision)

1.968 TFLOPS
🔥 1.754 TFLOPS

CUDA Cores

24,064
🔥 2,560

RT Cores

188
🔥 20

Architecture & Compatibility

GPU Architecture

Blackwell 2.0 Xe2

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

1,590
🔥 +68% 2,670

Boost Clock

2,617
🔥 +5% 2,740

Memory Clock

1,750
🔥 +36% 2,375

Power Consumption

TDP/TGP

600 W
🔥 -68% 190 W

Recommended PSU

1000 W
🔥 -55% 450 W

Power Connector

1x 16-pin 1x 8-pin

Rendering

Texture Units (TMU)

752
🔥 160

ROP

188
🔥 20

L2 Cache

128 MB
🔥 18 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 Dual-slot

Release Date

March 18, 2025 Dec. 13, 2024

Display Outputs

4x DisplayPort 2.1b
1x HDMI 2.1a
3x DisplayPort 2.1

Renting is cheaper than buying