can-i-run-this-llm

Can I run this LLM?

Can a Mac Studio M3 Ultra (96GB) run Muse Glimmer 30B?

Muse · Apache 2.0 · estimated for a 96.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Muse Glimmer 30B at Q8_0 · 29.6 GB · 17–25 tok/s on a Mac Studio M3 Ultra (96GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K11.0 GB38–58 tok/sRuns great
Q3_K_M14.0 GB32–48 tok/sRuns great
Q4_K_M17.3 GB27–40 tok/sRuns great
Q5_K_M20.1 GB24–35 tok/sRuns great
Q6_K23.4 GB21–31 tok/sRuns great
Q8_029.6 GB17–25 tok/sRuns great
FP1655.7 GB9–14 tok/sTight / slow

Ran Muse Glimmer 30B on a Mac Studio M3 Ultra (96GB)? Add your real numbers →