can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX 3090 (24GB) run Qwen3 30B-A3B (MoE)?

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

Yes — it runs great.

Best fit: Qwen3 30B-A3B (MoE) at Q3_K_M · 13.2 GB · 193–290 tok/s on a NVIDIA RTX 3090 (24GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K9.9 GB212–318 tok/sRuns great
Q3_K_M13.2 GB193–290 tok/sRuns great
Q4_K_M18.5 GB168–253 tok/sTight / slow
Q5_K_M21.8 GB156–234 tok/sWon't fit
Q6_K25.2 GB145–218 tok/sWon't fit
Q8_032.4 GB126–189 tok/sWon't fit
FP1661.1 GB83–125 tok/sWon't fit

Ran Qwen3 30B-A3B (MoE) on a NVIDIA RTX 3090 (24GB)? Add your real numbers →