can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA L40S (48GB) run Gemma 3 27B?

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

Yes — it runs great.

Best fit: Gemma 3 27B at Q8_0 · 28.7 GB · 20–29 tok/s on a NVIDIA L40S (48GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K10.5 GB45–67 tok/sRuns great
Q3_K_M13.4 GB37–55 tok/sRuns great
Q4_K_M16.6 GB31–47 tok/sRuns great
Q5_K_M19.3 GB28–41 tok/sRuns great
Q6_K22.2 GB25–37 tok/sRuns great
Q8_028.7 GB20–29 tok/sRuns great
FP1654.0 GB11–17 tok/sWon't fit

Ran Gemma 3 27B on a NVIDIA L40S (48GB)? Add your real numbers →