can it run
Yes, but close to the edge. 16GB against a 12GB working set: it runs, with little room for long context windows or other processes.
weights 9.1GB + KV cache 0.78GB at 4k context vs ~16GB usable → full. 9.9 GB needed of 16 GB usable — headroom for context
computed roofline: 732 GB/s × 0.17–0.38 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 | 16GB HBM2, 732 GB/s |
| params | 14B |
| license | mit |
| price | $135 |
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 Tesla P100 (16GB, used) can run — full list
→ tesla-v100-16gb — the next step up from this machine's GPU