Question 1

Can the AMD Ryzen AI Max (Strix Halo) run mistral-nemo 12B?

Accepted Answer

mistral-nemo 12B needs about 7.72GB working set (weights 7.1GB + KV cache 0.62GB at 4k context). The AMD Ryzen AI Max (Strix Halo) offers ~30GB usable — verdict: full. 7.7 GB needed of 30 GB usable — headroom for context

Question 2

How fast is mistral-nemo 12B on the AMD Ryzen AI Max (Strix Halo)?

Accepted Answer

~22–31 tok/s on the AMD Ryzen AI Max (Strix Halo). computed roofline: 256 GB/s × 0.60–0.85 efficiency window / 7.1GB (Q4_K_M) — bands, never points.

Question 3

What should I run mistral-nemo 12B on instead?

Accepted Answer

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

Can the AMD Ryzen AI Max (Strix Halo) run mistral-nemo 12B?

Run mistral-nemo 12B on instead