can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA H100 (80GB) run Gemma 4 31B?

Gemma · Apache 2.0 · estimated for a 80.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Gemma 4 31B at Q8_0 · 32.6 GB · 53–80 tok/s on a NVIDIA H100 (80GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M13.2 GB92–137 tok/sRuns great
Q4_K_M18.6 GB76–115 tok/sRuns great
Q5_K_M21.9 GB69–104 tok/sRuns great
Q6_K25.3 GB63–95 tok/sRuns great
Q8_032.6 GB53–80 tok/sRuns great
FP1661.4 GB33–49 tok/sTight / slow

Ran Gemma 4 31B on a NVIDIA H100 (80GB)? Add your real numbers →