can it run
14.2 GB needed of 22 GB usable — headroom for context
~22GB usable · NVIDIA RTX Spark N1X (Grace + Blackwell RTX) · model file 14GB · 20B MoE
computed roofline: 273 GB/s × 0.25–0.65 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
→ Tesla M40 (24GB, used) — cheapest catalog machine that runs gpt-oss 20B MoE fully ($50)
→ rtx-5090 — the next step up from NVIDIA RTX Spark N1X
→ every model the NVIDIA RTX Spark N1X can run — full list
→explore the full catalog97 machines indexed · live prices · what each one can run