can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA H100 SXM (80GB) run Phi-4 14B?

Phi · MIT · estimated for a 80.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Phi-4 14B at FP16 · 29.3 GB · 67–101 tok/s on a NVIDIA H100 SXM (80GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K5.6 GB188–281 tok/sRuns great
Q3_K_M7.2 GB168–251 tok/sRuns great
Q4_K_M8.9 GB150–225 tok/sRuns great
Q5_K_M10.4 GB138–206 tok/sRuns great
Q6_K12.0 GB126–189 tok/sRuns great
Q8_015.6 GB107–160 tok/sRuns great
FP1629.3 GB67–101 tok/sRuns great

Ran Phi-4 14B on a NVIDIA H100 SXM (80GB)? Add your real numbers →