can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX 3090 (24GB) run MiniCPM5 2B?

MiniCPM · Apache 2.0 · estimated for a 24.0 GB machine at 4K context.

Yes — it runs great.

Best fit: MiniCPM5 2B at FP16 · 5.0 GB · 97–146 tok/s on a NVIDIA RTX 3090 (24GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K1.0 GB225–338 tok/sRuns great
Q3_K_M1.2 GB209–314 tok/sRuns great
Q4_K_M1.6 GB188–282 tok/sRuns great
Q5_K_M1.9 GB174–261 tok/sRuns great
Q6_K2.1 GB166–249 tok/sRuns great
Q8_02.7 GB146–219 tok/sRuns great
FP165.0 GB97–146 tok/sRuns great
Open the full calculator →Everything a NVIDIA RTX 3090 (24GB) runsMiniCPM5 2B on other machinesWhat to buy to run it →

Ran MiniCPM5 2B on a NVIDIA RTX 3090 (24GB)? Add your real numbers →