local intelligence

can it run

Can the GTX 1660 Ti Mobile run llama3.1 8B?

TIGHT

5.4 of 6 GB usable — keep context modest

6GB · Turing TU116 · model file 4.9GB · 8B

~28–42 tok/s

computed roofline: 288 GB/s × 0.48–0.72 efficiency window / 4.9GB (Q4_K_M) — bands, never points.

Run llama3.1 8B on instead

Tesla P100 (16GB, used) — cheapest catalog machine that runs llama3.1 8B fully ($135)

every model the GTX 1660 Ti Mobile 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 1660 Ti Mobile · fit = weights + 4k KV vs usable memory