can-i-run-this-llm

Can I run this LLM?

Can a Intel Arc B580 (12GB) run Phi-4 14B?

Phi · MIT · estimated for a 12.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Phi-4 14B at Q3_K_M · 7.2 GB · 38–56 tok/s on a Intel Arc B580 (12GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K5.6 GB46–69 tok/sRuns great
Q3_K_M7.2 GB38–56 tok/sRuns great
Q4_K_M8.9 GB32–47 tok/sTight / slow
Q5_K_M10.4 GB28–42 tok/sWon't fit
Q6_K12.0 GB24–37 tok/sWon't fit
Q8_015.6 GB19–29 tok/sWon't fit
FP1629.3 GB11–16 tok/sWon't fit

Ran Phi-4 14B on a Intel Arc B580 (12GB)? Add your real numbers →