llamaperf

RTX 4090 vs RTX 5090

For running local LLMs · 13 reports across 2 models

Side A
RTX 4090
Vendor
nvidia
VRAM
24GB
Memory
Discrete
Side B
RTX 5090
Vendor
nvidia
VRAM
32GB
Memory
Discrete

Tokens per second by model

ModelRTX 4090RTX 5090
Qwen3.6up to 35B80.0n=23238.0n=7
Gemma 4up to 31B149.6n=1578.0n=3

More comparisons