Can I run this LLM?
Can a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB) run Gemma 4 12B?
Gemma · Apache 2.0 · estimated for a 128.0 GB machine at 4K context.
Yes — it runs great.
Best fit: Gemma 4 12B at Q6_K · 9.8 GB · 13–19 tok/s on a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q3_K_M | 5.1 GB | 19–29 tok/s | Runs great |
| Q4_K_M | 7.2 GB | 16–24 tok/s | Runs great |
| Q5_K_M | 8.5 GB | 14–21 tok/s | Runs great |
| Q6_K | 9.8 GB | 13–19 tok/s | Runs great |
| Q8_0 | 12.7 GB | 11–16 tok/s | Tight / slow |
| FP16 | 23.8 GB | 6–9 tok/s | Tight / slow |
Open the full calculator →Everything a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB) runsGemma 4 12B on other machinesWhat to buy to run it →
Ran Gemma 4 12B on a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB)? Add your real numbers →