can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA GH200 Grace Hopper (144GB HBM + 480GB) run Gemma 3 27B?

Gemma · Gemma · estimated for a 144.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Gemma 3 27B at FP16 · 54.0 GB · 55–82 tok/s on a NVIDIA GH200 Grace Hopper (144GB HBM + 480GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K10.5 GB161–242 tok/sRuns great
Q3_K_M13.4 GB143–214 tok/sRuns great
Q4_K_M16.6 GB127–190 tok/sRuns great
Q5_K_M19.3 GB116–174 tok/sRuns great
Q6_K22.2 GB106–159 tok/sRuns great
Q8_028.7 GB89–134 tok/sRuns great
FP1654.0 GB55–82 tok/sRuns great

Ran Gemma 3 27B on a NVIDIA GH200 Grace Hopper (144GB HBM + 480GB)? Add your real numbers →