can it run
No. 8GB of memory can't hold the 12GB working set this model needs. A smaller quant or a bigger machine is required.
weights 9.1GB + KV cache 0.78GB at 4k context vs ~8GB usable → offload-partial. 9.9 GB vs 8 GB usable — partial CPU offload, expect large speed loss
computed roofline: 448 GB/s × 0.48–0.72 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 | 8GB GDDR6, 448-496 GB/s |
| params | 14B |
| license | mit |
| price | $305 |
easy run: ollama pull phi4:14b
runs with: ollama · llama.cpp · koboldcpp · lm studio
→ Tesla P40 (24GB, used) — cheapest machine that runs it ($290)
→ every model the RTX 2080 / 2080 Super (8GB, used) can run — full list
→ rtx-3080-10gb — the next step up from this machine's GPU