Can I run this LLM?
Can a Mac Studio M5 Ultra (256GB) run Kimi Linear 48B-A3B (MoE)?
Kimi · MIT · estimated for a 256.0 GB machine at 4K context.
Yes — it runs great.
Best fit: Kimi Linear 48B-A3B (MoE) at FP16 · 98.3 GB · 76–114 tok/s on a Mac Studio M5 Ultra (256GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q2_K | 17.5 GB | 140–210 tok/s | Runs great |
| Q3_K_M | 22.7 GB | 133–199 tok/s | Runs great |
| Q4_K_M | 30.1 GB | 124–186 tok/s | Runs great |
| Q5_K_M | 35.1 GB | 118–177 tok/s | Runs great |
| Q6_K | 40.5 GB | 113–170 tok/s | Runs great |
| Q8_0 | 52.3 GB | 103–154 tok/s | Runs great |
| FP16 | 98.3 GB | 76–114 tok/s | Runs great |
Open the full calculator →Everything a Mac Studio M5 Ultra (256GB) runsKimi Linear 48B-A3B (MoE) on other machinesWhat to buy to run it →
Ran Kimi Linear 48B-A3B (MoE) on a Mac Studio M5 Ultra (256GB)? Add your real numbers →