can-i-run-this-llm

Can I run this LLM?

Can a Mac Studio M3 Ultra (512GB) run MiniCPM5 2B?

MiniCPM · Apache 2.0 · estimated for a 512.0 GB machine at 4K context.

Yes — it runs great.

Best fit: MiniCPM5 2B at FP16 · 5.0 GB · 65–98 tok/s on a Mac Studio M3 Ultra (512GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K1.0 GB125–188 tok/sRuns great
Q3_K_M1.2 GB119–179 tok/sRuns great
Q4_K_M1.6 GB110–165 tok/sRuns great
Q5_K_M1.9 GB104–156 tok/sRuns great
Q6_K2.1 GB100–151 tok/sRuns great
Q8_02.7 GB91–136 tok/sRuns great
FP165.0 GB65–98 tok/sRuns great

Ran MiniCPM5 2B on a Mac Studio M3 Ultra (512GB)? Add your real numbers →