Can I run this LLM?
Can a Mac Studio M1 Ultra (64GB) run Qwen3 Coder Next 80B-A3B (MoE)?
Qwen · Apache 2.0 · estimated for a 64.0 GB machine at 4K context.
Yes — it runs great.
Best fit: Qwen3 Coder Next 80B-A3B (MoE) at Q3_K_M · 38.3 GB · 108–162 tok/s on a Mac Studio M1 Ultra (64GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q2_K | 29.2 GB | 116–174 tok/s | Runs great |
| Q3_K_M | 38.3 GB | 108–162 tok/s | Runs great |
| Q4_K_M | 49.3 GB | 100–150 tok/s | Won't fit |
| Q5_K_M | 59.2 GB | 93–140 tok/s | Won't fit |
| Q6_K | 65.8 GB | 90–134 tok/s | Won't fit |
| Q8_0 | 84.8 GB | 80–120 tok/s | Won't fit |
| FP16 | 159.5 GB | 57–85 tok/s | Won't fit |
Open the full calculator →Everything a Mac Studio M1 Ultra (64GB) runsQwen3 Coder Next 80B-A3B (MoE) on other machinesWhat to buy to run it →
Ran Qwen3 Coder Next 80B-A3B (MoE) on a Mac Studio M1 Ultra (64GB)? Add your real numbers →