can-i-run-this-llm

Can I run this LLM?

Can a MacBook Air M4 (24GB) run Gemma 4 12B?

Gemma · Apache 2.0 · estimated for a 24.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Gemma 4 12B at Q3_K_M · 5.7 GB · 12–18 tok/s on a MacBook Air M4 (24GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M5.7 GB12–18 tok/sRuns great
Q4_K_M7.1 GB10–15 tok/sTight / slow
Q5_K_M8.4 GB9–13 tok/sTight / slow
Q6_K9.8 GB8–11 tok/sTight / slow
Q8_012.7 GB6–9 tok/sTight / slow
FP1623.8 GB3–5 tok/sWon't fit

Ran Gemma 4 12B on a MacBook Air M4 (24GB)? Add your real numbers →