Can I run this LLM?
Can a Mac Studio M3 Ultra (96GB) run Gemma 4 12B?
Gemma · Apache 2.0 · estimated for a 96.0 GB machine at 4K context.
Yes — it runs great.
Best fit: Gemma 4 12B at FP16 · 23.8 GB · 20–30 tok/s on a Mac Studio M3 Ultra (96GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q3_K_M | 5.7 GB | 59–89 tok/s | Runs great |
| Q4_K_M | 7.1 GB | 51–77 tok/s | Runs great |
| Q5_K_M | 8.4 GB | 46–69 tok/s | Runs great |
| Q6_K | 9.8 GB | 41–62 tok/s | Runs great |
| Q8_0 | 12.7 GB | 34–51 tok/s | Runs great |
| FP16 | 23.8 GB | 20–30 tok/s | Runs great |
Open the full calculator →Everything a Mac Studio M3 Ultra (96GB) runsGemma 4 12B on other machinesWhat to buy to run it →
Ran Gemma 4 12B on a Mac Studio M3 Ultra (96GB)? Add your real numbers →