can it run
Comfortably. 24GB of memory against a 4GB working set leaves headroom for context, the OS, and anything else you're running.
weights 2GB + KV cache 0.44GB at 4k context vs ~24GB usable → full. 2.4 GB needed of 24 GB usable — headroom for context
computed roofline: 672 GB/s × 0.48–0.72 efficiency window / 2.0GB (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 | 2.0GB (GGUF Q4_K_M) |
| minimum memory | 4GB working set |
| this machine | 24GB GDDR6, 672 GB/s |
| params | 3B |
| license | llama3.2 (gated on HF) |
| price | $900 |
easy run: ollama pull llama3.2:3b
runs with: ollama · llama.cpp · koboldcpp · lm studio
→ every model the Titan RTX (24GB, used) can run — full list
→ rtx-3090 — the next step up from this machine's GPU