GPU Wiki

Compare GPUs

Select up to four GPU Wiki entries to compare matching specifications row by row. The comparison uses the site database and keeps generation-specific specs visibly separate.

Popular comparisons

Add a third or fourth GPU
Spec RTX PRO 4000 Blackwell SFF
Overview
Product name NVIDIA RTX PRO 4000 Blackwell SFF
Architecture Blackwell 2.0
Generation / family NVIDIA RTX PRO Blackwell Series
GPU codename GB203
Launch date 2025-08-11
Market segment Desktop in NVIDIA RTX PRO Blackwell Series
Board type / entry type Workstation GPU
GPU
Process node TSMC 5 nm
Die size 378 mm²
Transistor count 45.6 billion
GPU chip GB203
SM / multiprocessor count 70
CUDA Cores 8960
Tensor Cores 280
RT Cores 70
ROPs 96
TMUs 280
Clocks
Base / Core clock 577 MHz
Boost clock 1,432 MHz
Memory clock 1,125 MHz
Memory
Memory size 24 GB
Memory type GDDR7
Memory bus width 192-bit
Memory bandwidth 432 GB/s
Bus / Interface
Bus interface PCIe 5.0 x8
AGP / PCI / PCIe distinction PCIe
Power
TDP / TBP / board power 70 W
Recommended PSU 250 W
Power connectors None
Slot width Dual-slot, 167 mm length, 40 mm width
Display / Media
Display outputs 4x mini-DisplayPort 2.1b
Features
DirectX support DirectX 12.2
OpenGL support OpenGL 4.6
Vulkan support Vulkan 1.4
CUDA support 12.0
Shader Model 6.8
Ray tracing support Hardware RT units present
Performance / Benchmarks
FP32 throughput 25.66 TFLOPS
FP16 throughput 25.66 TFLOPS
Muted cells indicate unavailable, not applicable, or not directly comparable specifications.

Frame rates compare runs of the same game at the same quality preset on one test system. DLAA counts as native resolution, because it applies anti-aliasing at full render resolution rather than upscaling; DLSS and FSR upscaling are never counted as native. Where a capture named ray tracing without a readable on or off value, it is read as on.