can-i-run-this-llm

Can I run this LLM?

Can a MacBook Pro M1 Pro (32GB) run Kimi Linear 48B-A3B (MoE)?

Kimi · MIT · estimated for a 32.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Kimi Linear 48B-A3B (MoE) at Q2_K · 16.0 GB · 74–111 tok/s on a MacBook Pro M1 Pro (32GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K16.0 GB74–111 tok/sRuns great
Q3_K_M21.2 GB63–94 tok/sTight / slow
Q4_K_M29.8 GB51–76 tok/sWon't fit
Q5_K_M35.0 GB45–68 tok/sWon't fit
Q6_K40.5 GB41–61 tok/sWon't fit
Q8_052.2 GB33–50 tok/sWon't fit
FP1698.2 GB20–30 tok/sWon't fit

Ran Kimi Linear 48B-A3B (MoE) on a MacBook Pro M1 Pro (32GB)? Add your real numbers →