Can I run this LLM?
Can a Mac Studio M3 Ultra (512GB) run Qwen3.8 Flash Next (MoE)?
Qwen · See model card · estimated for a 512.0 GB machine at 4K context.
Yes — it runs great.
Best fit: Qwen3.8 Flash Next (MoE) at Q8_0 · 188.3 GB · 57–85 tok/s on a Mac Studio M3 Ultra (512GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q2_K | 80.9 GB | 91–136 tok/s | Runs great |
| Q3_K_M | 92.0 GB | 86–128 tok/s | Runs great |
| Q4_K_M | 119.6 GB | 75–112 tok/s | Runs great |
| Q5_K_M | 134.7 GB | 70–105 tok/s | Runs great |
| Q6_K | 168.1 GB | 61–92 tok/s | Runs great |
| Q8_0 | 188.3 GB | 57–85 tok/s | Runs great |
| FP16 | 354.0 GB | 36–54 tok/s | Tight / slow |
Open the full calculator →Everything a Mac Studio M3 Ultra (512GB) runsQwen3.8 Flash Next (MoE) on other machinesWhat to buy to run it →
Ran Qwen3.8 Flash Next (MoE) on a Mac Studio M3 Ultra (512GB)? Add your real numbers →