Can I run this LLM?
Can a Mac Studio M5 Ultra (256GB) run Qwen3 30B-A3B (MoE)?
Qwen · Apache 2.0 · estimated for a 256.0 GB machine at 4K context.
Yes — it runs great.
Best fit: Qwen3 30B-A3B (MoE) at FP16 · 61.1 GB · 74–111 tok/s on a Mac Studio M5 Ultra (256GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q2_K | 11.3 GB | 131–196 tok/s | Runs great |
| Q3_K_M | 14.7 GB | 124–186 tok/s | Runs great |
| Q4_K_M | 18.6 GB | 117–176 tok/s | Runs great |
| Q5_K_M | 21.7 GB | 113–169 tok/s | Runs great |
| Q6_K | 25.1 GB | 108–162 tok/s | Runs great |
| Q8_0 | 32.5 GB | 98–148 tok/s | Runs great |
| FP16 | 61.1 GB | 74–111 tok/s | Runs great |
Open the full calculator →Everything a Mac Studio M5 Ultra (256GB) runsQwen3 30B-A3B (MoE) on other machinesWhat to buy to run it →
Ran Qwen3 30B-A3B (MoE) on a Mac Studio M5 Ultra (256GB)? Add your real numbers →