can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA L40S (48GB) run Phi-4 14B?

Phi · MIT · estimated for a 48.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Phi-4 14B at FP16 · 29.3 GB · 20–30 tok/s on a NVIDIA L40S (48GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K5.6 GB78–117 tok/sRuns great
Q3_K_M7.2 GB65–98 tok/sRuns great
Q4_K_M8.9 GB56–83 tok/sRuns great
Q5_K_M10.4 GB49–74 tok/sRuns great
Q6_K12.0 GB44–66 tok/sRuns great
Q8_015.6 GB35–53 tok/sRuns great
FP1629.3 GB20–30 tok/sRuns great

Ran Phi-4 14B on a NVIDIA L40S (48GB)? Add your real numbers →