can it run
19.0 GB vs 16 GB usable — partial CPU offload, expect large speed loss
16GB · Ada · model file 18.6GB · 30B-A3B MoE
computed roofline: 288 GB/s × 0.55–0.88 efficiency window / 1.9GB (Q4_K_M) — bands, never points.
MoE: file is 18.6GB (that decides fit) but only ~3B of experts activate per token (~1.9GB streamed) — band below is the active-expert estimate, optimistic; community MoE rows run several× the naive bandwidth math
→ Mac mini M4 (32GB) — cheapest catalog machine that runs qwen3 30B-A3B MoE fully ($999)
→ rtx-4070-ti-super — the next step up from NVIDIA GeForce RTX 4060 Ti 16GB
→ every model the NVIDIA GeForce RTX 4060 Ti 16GB can run — full list
→explore the full catalog84 machines indexed · live prices · what each one can run