can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA L40S (48GB) run Gemma 3 12B?

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

Yes — it runs great.

Best fit: Gemma 3 12B at FP16 · 23.5 GB · 24–36 tok/s on a NVIDIA L40S (48GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K4.8 GB79–118 tok/sRuns great
Q3_K_M6.0 GB68–102 tok/sRuns great
Q4_K_M7.3 GB60–90 tok/sRuns great
Q5_K_M8.5 GB54–81 tok/sRuns great
Q6_K9.7 GB49–74 tok/sRuns great
Q8_012.5 GB40–60 tok/sRuns great
FP1623.5 GB24–36 tok/sRuns great

Ran Gemma 3 12B on a NVIDIA L40S (48GB)? Add your real numbers →