can it run
9.9 GB vs 8 GB usable — partial CPU offload, expect large speed loss
8GB · Ampere · model file 9.1GB · 14B
computed roofline: 240 GB/s × 0.55–0.88 efficiency window / 9.1GB (Q4_K_M) — bands, never points.
→ Tesla P100 (16GB, used) — cheapest catalog machine that runs phi4 14B fully ($135)
→ rtx-3060-12 — the next step up from NVIDIA GeForce RTX 3060 8GB
→ every model the NVIDIA GeForce RTX 3060 8GB can run — full list
→explore the full catalog84 machines indexed · live prices · what each one can run