llamaperf

RTX 4070 vs RTX 5090

For running local LLMs · 6 reports across 2 models

Side A
RTX 4070
Vendor
nvidia
VRAM
12GB
Memory
Discrete
Side B
RTX 5090
Vendor
nvidia
VRAM
32GB
Memory
Discrete

Tokens per second by model

ModelRTX 4070RTX 5090
Gemma 4up to 31B55.0n=2n=1
Qwen3.6up to 35B106.5n=3

More comparisons