can-i-run-this-llm

Can I run this LLM?

Can a Mac Studio M2 Ultra (192GB) run Hunyuan Hy3 (MoE)?

Hunyuan · Apache 2.0 · estimated for a 192.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Hunyuan Hy3 (MoE) at Q2_K · 106.7 GB · 45–67 tok/s on a Mac Studio M2 Ultra (192GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K106.7 GB45–67 tok/sRuns great
Q3_K_M137.6 GB38–57 tok/sWon't fit
Q4_K_M182.2 GB31–47 tok/sWon't fit
Q5_K_M212.8 GB28–42 tok/sWon't fit
Q6_K257.3 GB24–36 tok/sWon't fit
Q8_0317.7 GB20–30 tok/sWon't fit

Ran Hunyuan Hy3 (MoE) on a Mac Studio M2 Ultra (192GB)? Add your real numbers →