can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX 4060 Ti (16GB) run Gemma 4 26B-A4B (MoE)?

Gemma · Apache 2.0 · estimated for a 16.0 GB machine at 4K context.

Yes, but it's tight.

Best fit: Gemma 4 26B-A4B (MoE) at Q3_K_M · 10.9 GB · 52–78 tok/s on a NVIDIA RTX 4060 Ti (16GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M10.9 GB52–78 tok/sTight / slow
Q4_K_M15.3 GB45–67 tok/sWon't fit
Q5_K_M18.0 GB41–61 tok/sWon't fit
Q6_K20.8 GB38–57 tok/sWon't fit
Q8_026.8 GB32–48 tok/sWon't fit
FP1650.5 GB21–31 tok/sWon't fit

Ran Gemma 4 26B-A4B (MoE) on a NVIDIA RTX 4060 Ti (16GB)? Add your real numbers →