Can I run this LLM?
Can a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB) run Gemma 4 E4B?
Gemma · Apache 2.0 · estimated for a 128.0 GB machine at 4K context.
Yes — it runs great.
Best fit: Gemma 4 E4B at FP16 · 15.1 GB · 17–26 tok/s on a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q3_K_M | 4.1 GB | 48–72 tok/s | Runs great |
| Q4_K_M | 5.0 GB | 42–63 tok/s | Runs great |
| Q5_K_M | 5.5 GB | 39–58 tok/s | Runs great |
| Q6_K | 7.1 GB | 32–48 tok/s | Runs great |
| Q8_0 | 8.2 GB | 29–43 tok/s | Runs great |
| FP16 | 15.1 GB | 17–26 tok/s | Runs great |
Open the full calculator →Everything a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB) runsGemma 4 E4B on other machinesWhat to buy to run it →
Ran Gemma 4 E4B on a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB)? Add your real numbers →