Question 1

Can the AMD Ryzen AI Max (Strix Halo) run qwen3 14B?

Accepted Answer

qwen3 14B needs about 9.92GB working set (weights 9.3GB + KV cache 0.62GB at 4k context). The AMD Ryzen AI Max (Strix Halo) offers ~30GB usable — verdict: full. 9.9 GB needed of 30 GB usable — headroom for context

Question 2

How fast is qwen3 14B on the AMD Ryzen AI Max (Strix Halo)?

Accepted Answer

~17–23 tok/s on the AMD Ryzen AI Max (Strix Halo). computed roofline: 256 GB/s × 0.60–0.85 efficiency window / 9.3GB (Q4_K_M) — bands, never points.

Question 3

What should I run qwen3 14B on instead?

Accepted Answer

Tesla P100 (16GB, used) runs it fully ($135).

Can the AMD Ryzen AI Max (Strix Halo) run qwen3 14B?

Run qwen3 14B on instead