can it run
11.1 GB vs 10 GB usable — partial CPU offload, expect large speed loss With a q8 (quantized) KV cache it comes to the edge (9.6GB) — possible with modest context, at some quality cost.
10GB · Battlemage · model file 8.0GB · 13B
computed roofline: 384 GB/s × 0.60–0.78 efficiency window / 8.0GB (Q4_K_M) — bands, never points.
→ Tesla P100 (16GB, used) — cheapest catalog machine that runs llava 13B fully ($135)
→ arc-b580 — the next step up from Intel Arc B570
→ every model the Intel Arc B570 can run — full list
→explore the full catalog84 machines indexed · live prices · what each one can run