Ling Tiny 3.0 is an 8‑billion‑parameter mixture‑of‑experts model with 1 billion active parameters that can be run via llama.cpp. The author executed it on a 2017 laptop with a 7th‑gen i5 CPU and 8 GB RAM, without any GPU or VRAM. In this low‑resource environment, the model produced a script to scan the local network for available llama.cpp servers.