NVIDIA GeForce RTX 5090 vs NVIDIA RTX PRO 6000 Blackwell Max-Q

Comparison of NVIDIA GeForce RTX 5090 with 32 GB GDDR7 and 21,760 cores vs NVIDIA RTX PRO 6000 Blackwell Max-Q with 96 GB GDDR7 and 24,064 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

NVIDIA GeForce RTX 5090

NVIDIA GeForce RTX 5090

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

NVIDIA RTX PRO 6000 Blackwell Max-Q

NVIDIA RTX PRO 6000 Blackwell Max-Q

Contents:

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

Memory

Memory Size

32 GB 96 GB

Memory Type

GDDR7 GDDR7

Memory Bandwidth

1.79 TB/s 1.79 TB/s

Memory Bus Width

512 бит 512 бит

ML Performance

FP16 (Half Precision)

104.8 TFLOPS 109.7 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

104.8 TFLOPS 109.7 TFLOPS

FP64 (Double Precision)

1.637 TFLOPS 1.715 TFLOPS

CUDA Cores

21,760 24,064

RT Cores

170 188

Architecture & Compatibility

GPU Architecture

Blackwell 2.0 Blackwell 2.0

SM (Streaming Multiprocessor)

170 188

PCIe Version

PCIe 5.0 x16 PCIe 5.0 x16

ML Software Support

CUDA Version

12.0 12.0

Clocks & Performance

Base Clock

2,017 1,035

Boost Clock

2,407 2,280

Memory Clock

1,750 1,750

Power Consumption

TDP/TGP

575 W
🔥 -48% 300 W

Recommended PSU

950 W
🔥 -26% 700 W

Power Connector

1x 16-pin 1x 16-pin

Rendering

Texture Units (TMU)

680 752

ROP

170 188

L2 Cache

96 MB 128 MB

Benchmarks

llama.cpp, llama 7B Q4_0

🔥 +17% 315.3 tokens/s
270.0 tokens/s

llama.cpp, llama-2-7b-Q4_0

273.7 tokens/s

Geekbench AI, FP16

🔥 +20% 69 765 points
58 178 points

Geekbench AI, INT8

🔥 +18% 34 585 points
29 188 points

Geekbench AI, FP32

🔥 +19% 46 834 points
39 339 points

Additional

Slots

Dual-slot Dual-slot

Release Date

Jan. 30, 2025 March 18, 2025

Display Outputs

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

Renting is cheaper than buying