can-i-run-this-llm

Can I run this LLM?

Can a Mac Studio M3 Ultra (512GB) run Kimi K2.7 (MoE)?

Kimi · See model card · estimated for a 512.0 GB machine at 4K context.

Yes, but it's tight.

Best fit: Kimi K2.7 (MoE) at Q2_K · 339.5 GB · 39–59 tok/s on a Mac Studio M3 Ultra (512GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K339.5 GB39–59 tok/sTight / slow
Q3_K_M463.6 GB31–46 tok/sWon't fit
Q4_K_M583.7 GB25–38 tok/sWon't fit

Ran Kimi K2.7 (MoE) on a Mac Studio M3 Ultra (512GB)? Add your real numbers →