can it run
Comfortably. 192GB of memory against a 19GB working set leaves headroom for context, the OS, and anything else you're running.
weights 14.6GB + KV cache 0.62GB at 4k context vs ~190GB usable → full. 15.2 GB needed of 190 GB usable — headroom for context
computed roofline: 256 GB/s × 0.60–0.85 efficiency window / 14.6GB (Q4_K_M) — bands, never points.
| model file | 14.6GB (GGUF Q4_K_M) |
| minimum memory | 19GB working set |
| this machine | up to 192GB LPDDR5X (160GB GPU-assignable) |
| params | 24B |
| license | apache-2.0 |
| price | $4500 |
easy run: ollama pull devstral:24b
runs with: ollama · llama.cpp · koboldcpp · lm studio
→ ASUS ExpertCenter Pro ET900N G3 (GB300) — this model on the biggest machine