RTX 3060 12GB vs RTX 3090

For running local LLMs · 6 reports across 3 models

Side A

Vendor: nvidia
VRAM: 12GB
Memory: Discrete

Side B

Vendor: nvidia
VRAM: 24GB
Memory: Discrete

Tokens per second by model

Model	RTX 3060 12GB	RTX 3090
Gemma 4up to 31B	60.0n=3	—
Qwen3.6up to 35B	—	66.0n=2
Qwen2.5	—	28.0n=1

More comparisons