can-i-run-this-llm

Can I run this LLM?

Can a AMD Instinct MI300X (192GB) run MiniCPM5 2B?

MiniCPM · Apache 2.0 · estimated for a 192.0 GB machine at 4K context.

Yes — it runs great.

Best fit: MiniCPM5 2B at FP16 · 5.0 GB · 246–370 tok/s on a AMD Instinct MI300X (192GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K1.0 GB330–495 tok/sRuns great
Q3_K_M1.2 GB324–486 tok/sRuns great
Q4_K_M1.6 GB314–471 tok/sRuns great
Q5_K_M1.9 GB307–460 tok/sRuns great
Q6_K2.1 GB302–453 tok/sRuns great
Q8_02.7 GB289–434 tok/sRuns great
FP165.0 GB246–370 tok/sRuns great

Ran MiniCPM5 2B on a AMD Instinct MI300X (192GB)? Add your real numbers →