can-i-run-this-llm

Can I run this LLM?

Can a Mac Studio M5 Ultra (256GB) run Mistral Small 4 119B (MoE)?

Mistral · Apache 2.0 · estimated for a 256.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Mistral Small 4 119B (MoE) at Q8_0 · 126.5 GB · 70–105 tok/s on a Mac Studio M5 Ultra (256GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K40.2 GB116–175 tok/sRuns great
Q3_K_M54.4 GB105–157 tok/sRuns great
Q4_K_M73.8 GB93–139 tok/sRuns great
Q5_K_M89.2 GB85–127 tok/sRuns great
Q6_K99.4 GB80–120 tok/sRuns great
Q8_0126.5 GB70–105 tok/sRuns great
FP16238.0 GB46–69 tok/sWon't fit

Ran Mistral Small 4 119B (MoE) on a Mac Studio M5 Ultra (256GB)? Add your real numbers →