can-i-run-this-llm

Can I run this LLM?

Can a Mac Studio M5 Ultra (96GB) run Gemma 3 12B?

Gemma · Gemma · estimated for a 96.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Gemma 3 12B at FP16 · 23.5 GB · 29–43 tok/s on a Mac Studio M5 Ultra (96GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K4.8 GB81–122 tok/sRuns great
Q3_K_M6.0 GB72–109 tok/sRuns great
Q4_K_M7.3 GB65–98 tok/sRuns great
Q5_K_M8.5 GB60–90 tok/sRuns great
Q6_K9.7 GB55–82 tok/sRuns great
Q8_012.5 GB46–69 tok/sRuns great
FP1623.5 GB29–43 tok/sRuns great

Ran Gemma 3 12B on a Mac Studio M5 Ultra (96GB)? Add your real numbers →