can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX Pro 6000 Blackwell (96GB) run Phi-4 14B?

Phi · MIT · estimated for a 96.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Phi-4 14B at FP16 · 29.3 GB · 39–59 tok/s on a NVIDIA RTX Pro 6000 Blackwell (96GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K5.6 GB132–197 tok/sRuns great
Q3_K_M7.2 GB114–171 tok/sRuns great
Q4_K_M8.9 GB99–149 tok/sRuns great
Q5_K_M10.4 GB89–134 tok/sRuns great
Q6_K12.0 GB80–120 tok/sRuns great
Q8_015.6 GB66–99 tok/sRuns great
FP1629.3 GB39–59 tok/sRuns great

Ran Phi-4 14B on a NVIDIA RTX Pro 6000 Blackwell (96GB)? Add your real numbers →