can-i-run-this-llm

Can I run this LLM?

Can a AMD Ryzen AI Max+ 395 “Strix Halo” (64GB) run Kimi Linear 48B-A3B (MoE)?

Kimi · MIT · estimated for a 64.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Kimi Linear 48B-A3B (MoE) at Q5_K_M · 35.1 GB · 54–81 tok/s on a AMD Ryzen AI Max+ 395 “Strix Halo” (64GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K17.5 GB82–123 tok/sRuns great
Q3_K_M22.7 GB71–107 tok/sRuns great
Q4_K_M30.1 GB60–90 tok/sRuns great
Q5_K_M35.1 GB54–81 tok/sRuns great
Q6_K40.5 GB49–74 tok/sTight / slow
Q8_052.3 GB41–61 tok/sWon't fit
FP1698.3 GB25–37 tok/sWon't fit

Ran Kimi Linear 48B-A3B (MoE) on a AMD Ryzen AI Max+ 395 “Strix Halo” (64GB)? Add your real numbers →