can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX 3060 Laptop (6GB) run MiniCPM5 2B?

MiniCPM · Apache 2.0 · estimated for a 6.0 GB machine at 4K context.

Yes — it runs great.

Best fit: MiniCPM5 2B at Q8_0 · 2.7 GB · 70–105 tok/s on a NVIDIA RTX 3060 Laptop (6GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K1.0 GB133–200 tok/sRuns great
Q3_K_M1.2 GB119–178 tok/sRuns great
Q4_K_M1.6 GB100–151 tok/sRuns great
Q5_K_M1.9 GB89–134 tok/sRuns great
Q6_K2.1 GB84–126 tok/sRuns great
Q8_02.7 GB70–105 tok/sRuns great
FP165.0 GB42–63 tok/sWon't fit

Ran MiniCPM5 2B on a NVIDIA RTX 3060 Laptop (6GB)? Add your real numbers →