can-i-run-this-llm

Can I run this LLM?

Can a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB) run Qwen3 30B-A3B (MoE)?

Qwen · Apache 2.0 · estimated for a 128.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Qwen3 30B-A3B (MoE) at FP16 · 61.1 GB · 23–35 tok/s on a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K9.9 GB72–109 tok/sRuns great
Q3_K_M13.2 GB64–96 tok/sRuns great
Q4_K_M18.5 GB53–80 tok/sRuns great
Q5_K_M21.8 GB49–73 tok/sRuns great
Q6_K25.2 GB44–67 tok/sRuns great
Q8_032.4 GB38–56 tok/sRuns great
FP1661.1 GB23–35 tok/sRuns great

Ran Qwen3 30B-A3B (MoE) on a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB)? Add your real numbers →