can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX Pro 6000 Blackwell (96GB) run Gemma 4 26B-A4B (MoE)?

Gemma · Apache 2.0 · estimated for a 96.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Gemma 4 26B-A4B (MoE) at FP16 · 50.5 GB · 111–167 tok/s on a NVIDIA RTX Pro 6000 Blackwell (96GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M12.7 GB224–336 tok/sRuns great
Q4_K_M17.0 GB201–301 tok/sRuns great
Q5_K_M21.2 GB183–274 tok/sRuns great
Q6_K23.2 GB175–262 tok/sRuns great
Q8_026.9 GB162–243 tok/sRuns great
FP1650.5 GB111–167 tok/sRuns great

Ran Gemma 4 26B-A4B (MoE) on a NVIDIA RTX Pro 6000 Blackwell (96GB)? Add your real numbers →