can-i-run-this-llm

Can I run this LLM?

Can a Mac Studio M5 Ultra (256GB) run Qwen3.8 Flash Next (MoE)?

Qwen · See model card · estimated for a 256.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Qwen3.8 Flash Next (MoE) at Q5_K_M · 134.7 GB · 87–131 tok/s on a Mac Studio M5 Ultra (256GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K80.9 GB108–162 tok/sRuns great
Q3_K_M92.0 GB103–154 tok/sRuns great
Q4_K_M119.6 GB92–138 tok/sRuns great
Q5_K_M134.7 GB87–131 tok/sRuns great
Q6_K168.1 GB78–117 tok/sTight / slow
Q8_0188.3 GB73–110 tok/sWon't fit
FP16354.0 GB49–73 tok/sWon't fit

Ran Qwen3.8 Flash Next (MoE) on a Mac Studio M5 Ultra (256GB)? Add your real numbers →