can-i-run-this-llm

Can I run this LLM?

Can a Mac Studio M3 Ultra (96GB) run Mistral Small 4 119B (MoE)?

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

Yes — it runs great.

Best fit: Mistral Small 4 119B (MoE) at Q3_K_M · 54.4 GB · 88–132 tok/s on a Mac Studio M3 Ultra (96GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K40.2 GB100–150 tok/sRuns great
Q3_K_M54.4 GB88–132 tok/sRuns great
Q4_K_M73.8 GB75–113 tok/sWon't fit
Q5_K_M89.2 GB68–101 tok/sWon't fit
Q6_K99.4 GB63–95 tok/sWon't fit
Q8_0126.5 GB54–81 tok/sWon't fit
FP16238.0 GB34–51 tok/sWon't fit

Ran Mistral Small 4 119B (MoE) on a Mac Studio M3 Ultra (96GB)? Add your real numbers →