Home NVIDIA RTX PRO 4500 Blackwell

NVIDIA RTX PRO 4500 Blackwell

This is a GPU manufactured with TSMC 5nm process, based on Nvidia Blackwell 2.0 architecture and released on Mar 2025. It features 45.6 billion transistors, 10496 CUDA cores and 32GB GDDR7 memory, with 64MB L2 cache, theoretical performance of 54.94TFLOPS, with total power consumption of 200W.

Graphics Card

[Report Issues]
Release Date
Mar 2025
Generation
Blackwell PRO W
Type
Desktop
Bus Interface
PCIe 5.0 x16

Clock Speeds

[Report Issues]
Base Clock
1590 MHz
Boost Clock
2617 MHz
Memory Clock
1750 MHz

Memory

[Report Issues]
Memory Size
32GB
Memory Type
GDDR7
Memory Bus
256bit
Bandwidth
896.0GB/s

Render Config

[Report Issues]
SM Count
82
Shading Units
10496
TMUs
328
ROPs
112
Tensor Cores
328
RT Cores
82
L1 Cache
128 KB (per SM)
L2 Cache
64 MB

Theoretical Performance

[Report Issues]
Pixel Rate
293.1 GPixel/s
Texture Rate
858.4 GTexel/s
FP16 (half)
54.94 TFLOPS
FP32 (float)
54.94 TFLOPS
FP64 (double)
858.4 GFLOPS

Graphics Processor

[Report Issues]
GPU Name
GB203
Architecture
Blackwell 2.0
Foundry
TSMC
Process Size
5 nm
Transistors
45.6 billion
Die Size
378 mm²

Board Design

[Report Issues]
TDP
200W
Suggested PSU
550 W
Outputs
4x DisplayPort 2.1b
Power Connectors
1x 16-pin

Graphics Features

[Report Issues]
DirectX
12 Ultimate (12_2)
OpenGL
4.6
OpenCL
3.0
Vulkan
1.4
CUDA
10.1
Shader Model
6.8

Rankings

[Report Issues]
FP32 (float)
NVIDIA H800 SXM5
NVIDIA H800 SXM5 80 GB HBM3
59.3 TFLOPS
NVIDIA GeForce RTX 5080
56.28 TFLOPS
NVIDIA GeForce RTX 5080 SUPER
56.28 TFLOPS
NVIDIA RTX PRO 4500 Blackwell
54.94 TFLOPS
NVIDIA H100 CNX
NVIDIA H100 CNX 80 GB HBM2e
53.84 TFLOPS
Intel Data Center GPU Max 1550
52.43 TFLOPS
Intel Data Center GPU Max Subsystem
52.43 TFLOPS
© 2025 - TopCPU.net