can-i-run-this-llm

Can I run this LLM?

Can a MacBook Pro M4 Max (128GB) run Gemma 4 31B?

Gemma · Apache 2.0 · estimated for a 128.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Gemma 4 31B at Q4_K_M · 18.6 GB · 13–20 tok/s on a MacBook Pro M4 Max (128GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M13.2 GB16–25 tok/sRuns great
Q4_K_M18.6 GB13–20 tok/sRuns great
Q5_K_M21.9 GB12–18 tok/sTight / slow
Q6_K25.3 GB11–16 tok/sTight / slow
Q8_032.6 GB9–13 tok/sTight / slow
FP1661.4 GB5–8 tok/sTight / slow

Ran Gemma 4 31B on a MacBook Pro M4 Max (128GB)? Add your real numbers →