can-i-run-this-llm

Can I run this LLM?

Can a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB) run Qwen3.6 27B?

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

Yes — it runs great.

Best fit: Qwen3.6 27B at Q3_K_M · 11.6 GB · 13–19 tok/s on a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M11.6 GB13–19 tok/sRuns great
Q4_K_M16.3 GB9–14 tok/sTight / slow
Q5_K_M19.2 GB8–12 tok/sTight / slow
Q6_K22.2 GB7–11 tok/sTight / slow
Q8_028.6 GB6–8 tok/sTight / slow
FP1653.8 GB3–5 tok/sTight / slow

Ran Qwen3.6 27B on a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB)? Add your real numbers →