can-i-run-this-llm

Can I run this LLM?

Can a Desktop CPU, DDR5 dual-channel (32GB) run Gemma 3 12B?

Gemma · Gemma · estimated for a 32.0 GB machine at 4K context.

Yes, but it's tight.

Best fit: Gemma 3 12B at Q2_K · 4.8 GB · 5–7 tok/s on a Desktop CPU, DDR5 dual-channel (32GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K4.8 GB5–7 tok/sTight / slow
Q3_K_M6.0 GB4–6 tok/sTight / slow
Q4_K_M7.3 GB3–5 tok/sTight / slow
Q5_K_M8.5 GB3–5 tok/sTight / slow
Q6_K9.7 GB3–4 tok/sTight / slow
Q8_012.5 GB2–3 tok/sTight / slow
FP1623.5 GB1–2 tok/sWon't fit

Ran Gemma 3 12B on a Desktop CPU, DDR5 dual-channel (32GB)? Add your real numbers →