can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX 4070 (12GB) run Gemma 4 12B?

Gemma · Apache 2.0 · estimated for a 12.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Gemma 4 12B at Q3_K_M · 5.1 GB · 42–63 tok/s on a NVIDIA RTX 4070 (12GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M5.1 GB42–63 tok/sRuns great
Q4_K_M7.2 GB34–51 tok/sTight / slow
Q5_K_M8.5 GB30–46 tok/sWon't fit
Q6_K9.8 GB27–41 tok/sWon't fit
Q8_012.7 GB23–34 tok/sWon't fit
FP1623.8 GB13–20 tok/sWon't fit

Ran Gemma 4 12B on a NVIDIA RTX 4070 (12GB)? Add your real numbers →