can-i-run-this-llm

Can I run this LLM?

Can a MacBook Pro M2 Max (64GB) run MiniCPM5 2B?

MiniCPM · Apache 2.0 · estimated for a 64.0 GB machine at 4K context.

Yes — it runs great.

Best fit: MiniCPM5 2B at FP16 · 5.0 GB · 40–60 tok/s on a MacBook Pro M2 Max (64GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K1.0 GB98–147 tok/sRuns great
Q3_K_M1.2 GB90–135 tok/sRuns great
Q4_K_M1.6 GB80–120 tok/sRuns great
Q5_K_M1.9 GB74–110 tok/sRuns great
Q6_K2.1 GB70–105 tok/sRuns great
Q8_02.7 GB61–91 tok/sRuns great
FP165.0 GB40–60 tok/sRuns great

Ran MiniCPM5 2B on a MacBook Pro M2 Max (64GB)? Add your real numbers →