can-i-run-this-llm

Can I run this LLM?

Can a AMD Instinct MI300X (192GB) run Gemma 4 31B?

Gemma · Apache 2.0 · estimated for a 192.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Gemma 4 31B at FP16 · 61.4 GB · 53–80 tok/s on a AMD Instinct MI300X (192GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M14.7 GB147–221 tok/sRuns great
Q4_K_M18.3 GB130–195 tok/sRuns great
Q5_K_M21.7 GB117–175 tok/sRuns great
Q6_K25.2 GB106–158 tok/sRuns great
Q8_032.6 GB88–132 tok/sRuns great
FP1661.4 GB53–80 tok/sRuns great

Ran Gemma 4 31B on a AMD Instinct MI300X (192GB)? Add your real numbers →