Can I run this LLM?
Can a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB) run Gemma 3 12B?
Gemma · Gemma · estimated for a 128.0 GB machine at 4K context.
Yes — it runs great.
Best fit: Gemma 3 12B at Q6_K · 9.7 GB · 14–21 tok/s on a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q2_K | 4.8 GB | 23–35 tok/s | Runs great |
| Q3_K_M | 6.0 GB | 20–30 tok/s | Runs great |
| Q4_K_M | 7.3 GB | 17–26 tok/s | Runs great |
| Q5_K_M | 8.5 GB | 16–23 tok/s | Runs great |
| Q6_K | 9.7 GB | 14–21 tok/s | Runs great |
| Q8_0 | 12.5 GB | 11–17 tok/s | Tight / slow |
| FP16 | 23.5 GB | 7–10 tok/s | Tight / slow |
Open the full calculator →Everything a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB) runsGemma 3 12B on other machinesWhat to buy to run it →
Ran Gemma 3 12B on a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB)? Add your real numbers →