can it run
Only with compromises. 11GB is below the 12GB working set — expect a lower quant, offloading, or heavy swap. Possible, not pleasant.
weights 9.1GB + KV cache 0.78GB at 4k context vs ~11GB usable → tight. 9.9 of 11 GB usable — keep context modest
computed roofline: 484 GB/s × 0.28–0.50 efficiency window / 9.1GB (Q4_K_M) — bands, never points.
| model file | 9.1GB (GGUF Q4_K_M) |
| minimum memory | 12GB working set |
| this machine | 11GB GDDR5X, 484 GB/s |
| params | 14B |
| license | mit |
| price | $275 |
easy run: ollama pull phi4:14b
runs with: ollama · llama.cpp · koboldcpp · lm studio
→ Tesla P40 (24GB, used) — cheapest machine that runs it ($290)
→ ASUS ExpertCenter Pro ET900N G3 (GB300) — this model on the biggest machine
→ every model the GTX 1080 Ti (11GB, used) can run — full list
→ RTX 3090 (24GB, used) — upgrade path machine