can it run
Only with compromises. 16GB is below the 18GB working set — expect a lower quant, offloading, or heavy swap. Possible, not pleasant.
weights 14GB + KV cache 0.19GB at 4k context vs ~16GB usable → tight. 14.2 of 16 GB usable — keep context modest
computed roofline: 160.4 GB/s × 0.28–0.50 efficiency window / 2.1GB (Q4_K_M) — bands, never points.
MoE: file is 14GB (that decides fit) but only ~3B of experts activate per token (~2.1GB streamed) — band below is the active-expert estimate, optimistic; community MoE rows run several× the naive bandwidth math
| model file | 14GB (GGUF Q4_K_M) |
| minimum memory | 18GB working set |
| this machine | 16GB GDDR5 (2×8 GB), 160 GB/s |
| params | 20B MoE |
| license | apache-2.0 |
| price | $35 |
easy run: ollama pull gpt-oss:20b
runs with: ollama · llama.cpp · koboldcpp · lm studio
→ Tesla V100 32GB (used) — cheapest machine that runs it ($645)
→ ASUS ExpertCenter Pro ET900N G3 (GB300) — this model on the biggest machine
→ every model the Tesla M60 (16GB, used) can run — full list
→ tesla-p100-16gb — the next step up from this machine's GPU