can-i-run-this-llm

Can I run this LLM?

Can a MacBook Pro M5 Max (36GB) run Gemma 4 26B-A4B (MoE)?

Gemma · Apache 2.0 · estimated for a 36.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Gemma 4 26B-A4B (MoE) at Q5_K_M · 21.2 GB · 57–85 tok/s on a MacBook Pro M5 Max (36GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M12.7 GB75–113 tok/sRuns great
Q4_K_M17.0 GB65–97 tok/sRuns great
Q5_K_M21.2 GB57–85 tok/sRuns great
Q6_K23.2 GB54–81 tok/sTight / slow
Q8_026.9 GB49–73 tok/sWon't fit
FP1650.5 GB31–46 tok/sWon't fit

Ran Gemma 4 26B-A4B (MoE) on a MacBook Pro M5 Max (36GB)? Add your real numbers →