can it run
Comfortably. 128GB of memory against a 4GB working set leaves headroom for context, the OS, and anything else you're running.
weights 1.7GB + KV cache 2GB at 4k context vs ~126GB usable → full. 3.7 GB needed of 126 GB usable — headroom for context
computed roofline: 256 GB/s × 0.60–0.85 efficiency window / 1.7GB (Q4_K_M) — bands, never points. Small models (under ~4.5B) run BELOW these bands — kernel-launch overhead keeps them from saturating bandwidth (measured against community medians, 09-06).
| model file | 1.7GB (GGUF Q4_K_M) |
| minimum memory | 4GB working set |
| this machine | 128GB LPDDR5X-8000 |
| params | 2B |
| license | apache-2.0 |
| price | $1999 |
easy run: ollama pull moondream:1.8b
runs with: ollama · llama.cpp · koboldcpp · lm studio
no alternatives — every machine in the catalog can run this one.