How much local AI can you squeeze out of a 16GB Mac Mini? Cloud is great, but I wanted to muck with the silicon. I vibe-coded a test harness to benchmark 13 local LLMs. From Llama 3.2's cache speed to the brutal 14B memory swap cliff, here is the data: blog.shiv.me/posts/2026/f...
From Digger to DeepSeek: A Quest for Local Silicon
Vibe-coding a 13-model benchmark harness to test the limits of local inference on a 16GB Mac Mini.
blog.shiv.me