Can I run this LLM?
Can a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB) run Qwen3.8 Flash Next (MoE)?
Qwen · See model card · estimated for a 128.0 GB machine at 4K context.
Yes, but it's tight.
Best fit: Qwen3.8 Flash Next (MoE) at Q2_K · 80.9 GB · 45–67 tok/s on a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q2_K | 80.9 GB | 45–67 tok/s | Tight / slow |
| Q3_K_M | 92.0 GB | 41–61 tok/s | Won't fit |
| Q4_K_M | 119.6 GB | 33–50 tok/s | Won't fit |
| Q5_K_M | 134.7 GB | 30–46 tok/s | Won't fit |
| Q6_K | 168.1 GB | 25–38 tok/s | Won't fit |
| Q8_0 | 188.3 GB | 23–35 tok/s | Won't fit |
| FP16 | 354.0 GB | 13–20 tok/s | Won't fit |
Open the full calculator →Everything a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB) runsQwen3.8 Flash Next (MoE) on other machinesWhat to buy to run it →
Ran Qwen3.8 Flash Next (MoE) on a AMD Ryzen AI Max+ 395 “Strix Halo” (128GB)? Add your real numbers →