can-i-run-this-llm

Can I run this LLM?

Can a MacBook Air/Pro M5 (32GB) run Gemma 4 26B-A4B (MoE)?

Gemma · Apache 2.0 · estimated for a 32.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Gemma 4 26B-A4B (MoE) at Q4_K_M · 17.0 GB · 29–43 tok/s on a MacBook Air/Pro M5 (32GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M12.7 GB35–53 tok/sRuns great
Q4_K_M17.0 GB29–43 tok/sRuns great
Q5_K_M21.2 GB24–36 tok/sTight / slow
Q6_K23.2 GB23–34 tok/sWon't fit
Q8_026.9 GB20–30 tok/sWon't fit
FP1650.5 GB12–17 tok/sWon't fit

Ran Gemma 4 26B-A4B (MoE) on a MacBook Air/Pro M5 (32GB)? Add your real numbers →