local intelligence

can it run

Can the GTX 965M run mixtral 8x7B MoE?

NO

26.5 GB vs 2 GB usable — does not fit

2GB · Maxwell (laptop) · model file 26GB · 8x7B MoE

~5.8–10 tok/s (active-expert estimate)

computed roofline: 80.19 GB/s × 0.28–0.50 efficiency window / 3.9GB (Q4_K_M) — bands, never points.

MoE: file is 26GB (that decides fit) but only ~1.05B of experts activate per token (~3.9GB streamed) — band below is the active-expert estimate, optimistic; community MoE rows run several× the naive bandwidth math

Run mixtral 8x7B MoE on instead

Mac mini M5 Pro (64GB) — cheapest catalog machine that runs mixtral 8x7B MoE fully ($1669)

every model the GTX 965M can run — full list

explore the full catalog84 machines indexed · live prices · what each one can run
2026-09-17 · ← all models on the GTX 965M · fit = weights + 4k KV vs usable memory