can-i-run-this-llm

Can I run this LLM?

Can a Mac Studio M3 Ultra (96GB) run Gemma 3 27B?

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

Yes — it runs great.

Best fit: Gemma 3 27B at Q8_0 · 28.7 GB · 16–24 tok/s on a Mac Studio M3 Ultra (96GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K10.5 GB35–52 tok/sRuns great
Q3_K_M13.4 GB29–44 tok/sRuns great
Q4_K_M16.6 GB25–38 tok/sRuns great
Q5_K_M19.3 GB22–34 tok/sRuns great
Q6_K22.2 GB20–30 tok/sRuns great
Q8_028.7 GB16–24 tok/sRuns great
FP1654.0 GB9–14 tok/sTight / slow

Ran Gemma 3 27B on a Mac Studio M3 Ultra (96GB)? Add your real numbers →