can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX 4090 (24GB) run Gemma 4 26B-A4B (MoE)?

Gemma · Apache 2.0 · estimated for a 24.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Gemma 4 26B-A4B (MoE) at Q4_K_M · 17.0 GB · 149–223 tok/s on a NVIDIA RTX 4090 (24GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M12.7 GB172–257 tok/sRuns great
Q4_K_M17.0 GB149–223 tok/sRuns great
Q5_K_M21.2 GB131–197 tok/sWon't fit
Q6_K23.2 GB124–186 tok/sWon't fit
Q8_026.9 GB113–170 tok/sWon't fit
FP1650.5 GB72–108 tok/sWon't fit

Ran Gemma 4 26B-A4B (MoE) on a NVIDIA RTX 4090 (24GB)? Add your real numbers →