Can I run this LLM?
Can a AMD Ryzen AI Max+ 395 “Strix Halo” (64GB) run Qwen3.8 27B?
Qwen · Apache 2.0 · estimated for a 64.0 GB machine at 4K context.
Yes, but it's tight.
Best fit: Qwen3.8 27B at Q3_K_M · 13.8 GB · 11–16 tok/s on a AMD Ryzen AI Max+ 395 “Strix Halo” (64GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q3_K_M | 13.8 GB | 11–16 tok/s | Tight / slow |
| Q4_K_M | 17.1 GB | 9–13 tok/s | Tight / slow |
| Q5_K_M | 19.8 GB | 8–12 tok/s | Tight / slow |
| Q6_K | 22.9 GB | 7–10 tok/s | Tight / slow |
| Q8_0 | 29.1 GB | 6–8 tok/s | Tight / slow |
| FP16 | 54.7 GB | 3–5 tok/s | Won't fit |
Open the full calculator →Everything a AMD Ryzen AI Max+ 395 “Strix Halo” (64GB) runsQwen3.8 27B on other machinesWhat to buy to run it →
Ran Qwen3.8 27B on a AMD Ryzen AI Max+ 395 “Strix Halo” (64GB)? Add your real numbers →