can-i-run-this-llm

Can I run this LLM?

Can a AMD Instinct MI300X (192GB) run Gemma 4 12B?

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

Yes — it runs great.

Best fit: Gemma 4 12B at FP16 · 23.8 GB · 105–158 tok/s on a AMD Instinct MI300X (192GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M5.1 GB211–316 tok/sRuns great
Q4_K_M7.2 GB190–284 tok/sRuns great
Q5_K_M8.5 GB179–268 tok/sRuns great
Q6_K9.8 GB168–253 tok/sRuns great
Q8_012.7 GB150–225 tok/sRuns great
FP1623.8 GB105–158 tok/sRuns great

Ran Gemma 4 12B on a AMD Instinct MI300X (192GB)? Add your real numbers →